[{"data":1,"prerenderedAt":18885},["ShallowReactive",2],{"lesson:\u002Fnatural-language-processing\u002Fsequences\u002Frnns-and-lstms":3,"course-wordcounts":13138,"ref-card-index":14050,"nav:natural-language-processing":18731,"tikz:96f30c51fe62e45e12efc9f46a062d97da62a477bbe34cd92ccd6ab67fce29c4":18877,"tikz:9bf56bc9c0cb8c460e6340888df2136a9d399c097ead1ae79b100ff814e7bd56":18878,"tikz:295cce872b423963a7529b2952ca9e7d33ae697b06f5c0aabeed6611d43da8b3":18879,"tikz:98d0012c7dfa26b76a946676e25837d1c65d9435fc3dbc601a51455a14552017":18880,"tikz:4af20d242d8806800bf5557f3d5439c8d73f20c1ed7434f8e303dc6830e8c45e":18881,"tikz:197f74586cadd86b53416051d5e03353ce98ce76de6e6079f5c72f49d98e1d5b":18882,"tikz:0288c11dbf62fd8adec35ef80e7ace60b0d19a96d211e07d83b7e9a5bceaed1b":18883,"tikz:c75307818047ae227237d25d50490fe7716c22943f06205cc2ef9458bffd42d6":18884},{"id":4,"title":5,"blurb":6,"body":7,"brief":13109,"category":13110,"description":13111,"draft":13112,"extension":13113,"meta":13114,"module":13116,"navigation":13117,"path":13118,"practice":13119,"rawbody":13120,"readingTime":13121,"seo":13126,"sources":13127,"status":13133,"stem":13134,"summary":13135,"topics":13136,"__hash__":13137},"course\u002F09.natural-language-processing\u002F04.sequences\u002F03.rnns-and-lstms.md","RNNs and LSTMs","",{"type":8,"value":9,"toc":13088},"minimark",[10,106,126,131,298,525,613,766,1318,1322,1436,1439,1962,1967,2085,2137,2140,2144,2469,2813,3386,3390,3454,3490,3946,4310,4786,4958,4962,5044,5423,5580,5583,5647,5719,5723,5972,6021,6025,6363,7270,7849,7852,7884,7888,8006,8025,8028,8058,8062,8137,8271,8395,8852,8874,9152,9158,9657,9663,10116,10119,10411,10489,10493,10570,10685,10978,11099,11719,11731,11735,11750,11754,11757,11790,11809,11837,11841,11844,11851,11905,12585,12657,12660,12664,12686,12701,12704,12731,12745,13084],[11,12,13,14,19,20,24,25,75,76,92,93,101,102,105],"p",{},"The ",[15,16,18],"a",{"href":17},"\u002Fnatural-language-processing\u002Fsemantics\u002Fneural-language-models","feedforward neural language model","\nfixed the sparsity of n-grams by embedding words as vectors, but it kept one limitation\nfrom them: a ",[21,22,23],"strong",{},"fixed window",". It reads the last ",[26,27,30],"span",{"className":28},[29],"katex",[26,31,35,64],{"className":32,"ariaHidden":34},[33],"katex-html","true",[26,36,39,44,51,56,61],{"className":37},[38],"base",[26,40],{"className":41,"style":43},[42],"strut","height:0.7667em;vertical-align:-0.0833em;",[26,45,50],{"className":46,"style":49},[47,48],"mord","mathnormal","margin-right:0.109em;","N",[26,52],{"className":53,"style":55},[54],"mspace","margin-right:0.2222em;",[26,57,60],{"className":58},[59],"mbin","−",[26,62],{"className":63,"style":55},[54],[26,65,67,71],{"className":66},[38],[26,68],{"className":69,"style":70},[42],"height:0.6444em;",[26,72,74],{"className":73},[47],"1"," words, concatenates\ntheir embeddings, and predicts the next word — and a word ",[26,77,79],{"className":78},[29],[26,80,82],{"className":81,"ariaHidden":34},[33],[26,83,85,89],{"className":84},[38],[26,86],{"className":87,"style":88},[42],"height:0.6833em;",[26,90,50],{"className":91,"style":49},[47,48]," positions back is simply\ninvisible. Widen the window and the input layer grows with it; the model still cannot\ncondition on a subject that appeared twenty words ago. Linguistic dependencies span\narbitrary distances:\nthe verb in ",[94,95,96,97,100],"em",{},"the keys to the cabinet ",[21,98,99],{},"are"," on the table"," agrees with ",[94,103,104],{},"keys",", five words\nback, across a singular noun that would mislead any fixed-window guess.",[11,107,13,108,111,112,115,116,125],{},[21,109,110],{},"recurrent neural network"," removes the window entirely. Instead of reading a fixed\nwindow of recent words, it processes the sequence one token at a time and carries a ",[21,113,114],{},"hidden\nstate"," forward from step to step. That state is a running summary of everything read so\nfar, with no built-in limit on how far back it can reach.",[117,118,119],"sup",{},[15,120,74],{"href":121,"ariaDescribedBy":122,"dataFootnoteRef":6,"id":124},"#user-content-fn-jm-rnn",[123],"footnote-label","user-content-fnref-jm-rnn"," This lesson builds the\nRNN from its one recurrent equation, turns it into a language model, trains it, finds the\ngradient problem that undermines it, and repairs that problem with the LSTM.",[127,128,130],"h2",{"id":129},"the-recurrent-equation","The recurrent equation",[11,132,133,134,137,138,141,142,159,160,231,232,297],{},"A recurrent network is any network with a cycle in its connections: the value of some unit\ndepends, directly or indirectly, on its own earlier output. The constrained, well-behaved\nversion used throughout language processing is the ",[21,135,136],{},"Elman network",", or ",[21,139,140],{},"simple recurrent\nnetwork",". At each time step ",[26,143,145],{"className":144},[29],[26,146,148],{"className":147,"ariaHidden":34},[33],[26,149,151,155],{"className":150},[38],[26,152],{"className":153,"style":154},[42],"height:0.6151em;",[26,156,158],{"className":157},[47,48],"t"," it takes the current input vector ",[26,161,163],{"className":162},[29],[26,164,166],{"className":165,"ariaHidden":34},[33],[26,167,169,173],{"className":168},[38],[26,170],{"className":171,"style":172},[42],"height:0.5944em;vertical-align:-0.15em;",[26,174,176,181],{"className":175},[47],[26,177,180],{"className":178},[47,179],"mathbf","x",[26,182,185],{"className":183},[184],"msupsub",[26,186,190,222],{"className":187},[188,189],"vlist-t","vlist-t2",[26,191,194,217],{"className":192},[193],"vlist-r",[26,195,199],{"className":196,"style":198},[197],"vlist","height:0.2806em;",[26,200,202,207],{"style":201},"top:-2.55em;margin-left:0em;margin-right:0.05em;",[26,203],{"className":204,"style":206},[205],"pstrut","height:2.7em;",[26,208,214],{"className":209},[210,211,212,213],"sizing","reset-size6","size3","mtight",[26,215,158],{"className":216},[47,48,213],[26,218,221],{"className":219},[220],"vlist-s","​",[26,223,225],{"className":224},[193],[26,226,229],{"className":227,"style":228},[197],"height:0.15em;",[26,230],{}," and the\nhidden state ",[26,233,235],{"className":234},[29],[26,236,238],{"className":237,"ariaHidden":34},[33],[26,239,241,245],{"className":240},[38],[26,242],{"className":243,"style":244},[42],"height:0.9028em;vertical-align:-0.2083em;",[26,246,248,252],{"className":247},[47],[26,249,251],{"className":250},[47,179],"h",[26,253,255],{"className":254},[184],[26,256,258,288],{"className":257},[188,189],[26,259,261,285],{"className":260},[193],[26,262,265],{"className":263,"style":264},[197],"height:0.3011em;",[26,266,267,270],{"style":201},[26,268],{"className":269,"style":206},[205],[26,271,273],{"className":272},[210,211,212,213],[26,274,276,279,282],{"className":275},[47,213],[26,277,158],{"className":278},[47,48,213],[26,280,60],{"className":281},[59,213],[26,283,74],{"className":284},[47,213],[26,286,221],{"className":287},[220],[26,289,291],{"className":290},[193],[26,292,295],{"className":293,"style":294},[197],"height:0.2083em;",[26,296],{}," from the previous step, and produces a new hidden state:",[26,299,302],{"className":300},[301],"katex-display",[26,303,305],{"className":304},[29],[26,306,308,373],{"className":307,"ariaHidden":34},[33],[26,309,311,315,355,359,362,367,370],{"className":310},[38],[26,312],{"className":313,"style":314},[42],"height:0.8444em;vertical-align:-0.15em;",[26,316,318,321],{"className":317},[47],[26,319,251],{"className":320},[47,179],[26,322,324],{"className":323},[184],[26,325,327,347],{"className":326},[188,189],[26,328,330,344],{"className":329},[193],[26,331,333],{"className":332,"style":198},[197],[26,334,335,338],{"style":201},[26,336],{"className":337,"style":206},[205],[26,339,341],{"className":340},[210,211,212,213],[26,342,158],{"className":343},[47,48,213],[26,345,221],{"className":346},[220],[26,348,350],{"className":349},[193],[26,351,353],{"className":352,"style":228},[197],[26,354],{},[26,356],{"className":357,"style":358},[54],"margin-right:0.2778em;",[26,360],{"className":361,"style":358},[54],[26,363,366],{"className":364},[365],"mrel","=",[26,368],{"className":369,"style":358},[54],[26,371],{"className":372,"style":358},[54],[26,374,376,380,385,389,393,517,520],{"className":375},[38],[26,377],{"className":378,"style":379},[42],"height:1em;vertical-align:-0.25em;",[26,381,384],{"className":382,"style":383},[47,48],"margin-right:0.0359em;","g",[26,386],{"className":387,"style":388},[54],"margin-right:-0.1667em;",[26,390],{"className":391,"style":392},[54],"margin-right:0.1667em;",[26,394,397,404,408,457,460,464,467,472,512],{"className":395},[396],"minner",[26,398,403],{"className":399,"style":402},[400,401],"mopen","delimcenter","top:0em;","(",[26,405,407],{"className":406},[47,179],"U",[26,409,411,414],{"className":410},[47],[26,412,251],{"className":413},[47,179],[26,415,417],{"className":416},[184],[26,418,420,449],{"className":419},[188,189],[26,421,423,446],{"className":422},[193],[26,424,426],{"className":425,"style":264},[197],[26,427,428,431],{"style":201},[26,429],{"className":430,"style":206},[205],[26,432,434],{"className":433},[210,211,212,213],[26,435,437,440,443],{"className":436},[47,213],[26,438,158],{"className":439},[47,48,213],[26,441,60],{"className":442},[59,213],[26,444,74],{"className":445},[47,213],[26,447,221],{"className":448},[220],[26,450,452],{"className":451},[193],[26,453,455],{"className":454,"style":294},[197],[26,456],{},[26,458],{"className":459,"style":55},[54],[26,461,463],{"className":462},[59],"+",[26,465],{"className":466,"style":55},[54],[26,468,471],{"className":469,"style":470},[47,179],"margin-right:0.016em;","W",[26,473,475,478],{"className":474},[47],[26,476,180],{"className":477},[47,179],[26,479,481],{"className":480},[184],[26,482,484,504],{"className":483},[188,189],[26,485,487,501],{"className":486},[193],[26,488,490],{"className":489,"style":198},[197],[26,491,492,495],{"style":201},[26,493],{"className":494,"style":206},[205],[26,496,498],{"className":497},[210,211,212,213],[26,499,158],{"className":500},[47,48,213],[26,502,221],{"className":503},[220],[26,505,507],{"className":506},[193],[26,508,510],{"className":509,"style":228},[197],[26,511],{},[26,513,516],{"className":514,"style":402},[515,401],"mclose",")",[26,518],{"className":519,"style":392},[54],[26,521,524],{"className":522},[523],"mpunct",",",[11,526,527,528,544,545,563,564,580,581,596,597,612],{},"where ",[26,529,531],{"className":530},[29],[26,532,534],{"className":533,"ariaHidden":34},[33],[26,535,537,541],{"className":536},[38],[26,538],{"className":539,"style":540},[42],"height:0.625em;vertical-align:-0.1944em;",[26,542,384],{"className":543,"style":383},[47,48]," is a non-linearity (typically ",[26,546,548],{"className":547},[29],[26,549,551],{"className":550,"ariaHidden":34},[33],[26,552,554,558],{"className":553},[38],[26,555],{"className":556,"style":557},[42],"height:0.6944em;",[26,559,562],{"className":560},[561],"mop","tanh","), ",[26,565,567],{"className":566},[29],[26,568,570],{"className":569,"ariaHidden":34},[33],[26,571,573,577],{"className":572},[38],[26,574],{"className":575,"style":576},[42],"height:0.6861em;",[26,578,471],{"className":579,"style":470},[47,179]," maps the input into the\nhidden layer, and ",[26,582,584],{"className":583},[29],[26,585,587],{"className":586,"ariaHidden":34},[33],[26,588,590,593],{"className":589},[38],[26,591],{"className":592,"style":576},[42],[26,594,407],{"className":595},[47,179]," maps the previous hidden state into the current one. The\nrecurrent matrix ",[26,598,600],{"className":599},[29],[26,601,603],{"className":602,"ariaHidden":34},[33],[26,604,606,609],{"className":605},[38],[26,607],{"className":608,"style":576},[42],[26,610,407],{"className":611},[47,179]," is the whole difference from a feedforward network: it is how\npast context flows into the present. From the hidden state we read an output in the usual\nway, a softmax over classes for soft classification:",[26,614,616],{"className":615},[301],[26,617,619],{"className":618},[29],[26,620,622,686],{"className":621,"ariaHidden":34},[33],[26,623,625,629,671,674,677,680,683],{"className":624},[38],[26,626],{"className":627,"style":628},[42],"height:0.6389em;vertical-align:-0.1944em;",[26,630,632,636],{"className":631},[47],[26,633,635],{"className":634,"style":470},[47,179],"y",[26,637,639],{"className":638},[184],[26,640,642,663],{"className":641},[188,189],[26,643,645,660],{"className":644},[193],[26,646,648],{"className":647,"style":198},[197],[26,649,651,654],{"style":650},"top:-2.55em;margin-left:-0.016em;margin-right:0.05em;",[26,652],{"className":653,"style":206},[205],[26,655,657],{"className":656},[210,211,212,213],[26,658,158],{"className":659},[47,48,213],[26,661,221],{"className":662},[220],[26,664,666],{"className":665},[193],[26,667,669],{"className":668,"style":228},[197],[26,670],{},[26,672],{"className":673,"style":358},[54],[26,675],{"className":676,"style":358},[54],[26,678,366],{"className":679},[365],[26,681],{"className":682,"style":358},[54],[26,684],{"className":685,"style":358},[54],[26,687,689,692,700,703,706,759,762],{"className":688},[38],[26,690],{"className":691,"style":379},[42],[26,693,695],{"className":694},[47],[26,696,699],{"className":697},[47,698],"mathrm","softmax",[26,701],{"className":702,"style":388},[54],[26,704],{"className":705,"style":392},[54],[26,707,709,712,716,756],{"className":708},[396],[26,710,403],{"className":711,"style":402},[400,401],[26,713,715],{"className":714,"style":470},[47,179],"V",[26,717,719,722],{"className":718},[47],[26,720,251],{"className":721},[47,179],[26,723,725],{"className":724},[184],[26,726,728,748],{"className":727},[188,189],[26,729,731,745],{"className":730},[193],[26,732,734],{"className":733,"style":198},[197],[26,735,736,739],{"style":201},[26,737],{"className":738,"style":206},[205],[26,740,742],{"className":741},[210,211,212,213],[26,743,158],{"className":744},[47,48,213],[26,746,221],{"className":747},[220],[26,749,751],{"className":750},[193],[26,752,754],{"className":753,"style":228},[197],[26,755],{},[26,757,516],{"className":758,"style":402},[515,401],[26,760],{"className":761,"style":392},[54],[26,763,765],{"className":764},[47],".",[11,767,768,769,932,933,1075,1076,1230,1231,1234,1235,1250,1251,1266,1267,1283,1284,1300,1301,1317],{},"Three weight matrices define the network — ",[26,770,772],{"className":771},[29],[26,773,775,795],{"className":774,"ariaHidden":34},[33],[26,776,778,782,785,788,792],{"className":777},[38],[26,779],{"className":780,"style":781},[42],"height:0.7252em;vertical-align:-0.0391em;",[26,783,471],{"className":784,"style":470},[47,179],[26,786],{"className":787,"style":358},[54],[26,789,791],{"className":790},[365],"∈",[26,793],{"className":794,"style":358},[54],[26,796,798,802],{"className":797},[38],[26,799],{"className":800,"style":801},[42],"height:0.8491em;",[26,803,805,810],{"className":804},[47],[26,806,809],{"className":807},[47,808],"mathbb","R",[26,811,813],{"className":812},[184],[26,814,816],{"className":815},[188],[26,817,819],{"className":818},[193],[26,820,822],{"className":821,"style":801},[197],[26,823,825,828],{"style":824},"top:-3.063em;margin-right:0.05em;",[26,826],{"className":827,"style":206},[205],[26,829,831],{"className":830},[210,211,212,213],[26,832,834,881,885],{"className":833},[47,213],[26,835,837,841],{"className":836},[47,213],[26,838,840],{"className":839},[47,48,213],"d",[26,842,844],{"className":843},[184],[26,845,847,872],{"className":846},[188,189],[26,848,850,869],{"className":849},[193],[26,851,854],{"className":852,"style":853},[197],"height:0.3448em;",[26,855,857,861],{"style":856},"top:-2.3488em;margin-left:0em;margin-right:0.0714em;",[26,858],{"className":859,"style":860},[205],"height:2.5em;",[26,862,866],{"className":863},[210,864,865,213],"reset-size3","size1",[26,867,251],{"className":868},[47,48,213],[26,870,221],{"className":871},[220],[26,873,875],{"className":874},[193],[26,876,879],{"className":877,"style":878},[197],"height:0.1512em;",[26,880],{},[26,882,884],{"className":883},[59,213],"×",[26,886,888,891],{"className":887},[47,213],[26,889,840],{"className":890},[47,48,213],[26,892,894],{"className":893},[184],[26,895,897,923],{"className":896},[188,189],[26,898,900,920],{"className":899},[193],[26,901,904],{"className":902,"style":903},[197],"height:0.3281em;",[26,905,907,910],{"style":906},"top:-2.357em;margin-left:0em;margin-right:0.0714em;",[26,908],{"className":909,"style":860},[205],[26,911,913],{"className":912},[210,864,865,213],[26,914,916],{"className":915},[47,213],[26,917,919],{"className":918},[47,48,213],"in",[26,921,221],{"className":922},[220],[26,924,926],{"className":925},[193],[26,927,930],{"className":928,"style":929},[197],"height:0.143em;",[26,931],{},",\n",[26,934,936],{"className":935},[29],[26,937,939,957],{"className":938,"ariaHidden":34},[33],[26,940,942,945,948,951,954],{"className":941},[38],[26,943],{"className":944,"style":781},[42],[26,946,407],{"className":947},[47,179],[26,949],{"className":950,"style":358},[54],[26,952,791],{"className":953},[365],[26,955],{"className":956,"style":358},[54],[26,958,960,963],{"className":959},[38],[26,961],{"className":962,"style":801},[42],[26,964,966,969],{"className":965},[47],[26,967,809],{"className":968},[47,808],[26,970,972],{"className":971},[184],[26,973,975],{"className":974},[188],[26,976,978],{"className":977},[193],[26,979,981],{"className":980,"style":801},[197],[26,982,983,986],{"style":824},[26,984],{"className":985,"style":206},[205],[26,987,989],{"className":988},[210,211,212,213],[26,990,992,1032,1035],{"className":991},[47,213],[26,993,995,998],{"className":994},[47,213],[26,996,840],{"className":997},[47,48,213],[26,999,1001],{"className":1000},[184],[26,1002,1004,1024],{"className":1003},[188,189],[26,1005,1007,1021],{"className":1006},[193],[26,1008,1010],{"className":1009,"style":853},[197],[26,1011,1012,1015],{"style":856},[26,1013],{"className":1014,"style":860},[205],[26,1016,1018],{"className":1017},[210,864,865,213],[26,1019,251],{"className":1020},[47,48,213],[26,1022,221],{"className":1023},[220],[26,1025,1027],{"className":1026},[193],[26,1028,1030],{"className":1029,"style":878},[197],[26,1031],{},[26,1033,884],{"className":1034},[59,213],[26,1036,1038,1041],{"className":1037},[47,213],[26,1039,840],{"className":1040},[47,48,213],[26,1042,1044],{"className":1043},[184],[26,1045,1047,1067],{"className":1046},[188,189],[26,1048,1050,1064],{"className":1049},[193],[26,1051,1053],{"className":1052,"style":853},[197],[26,1054,1055,1058],{"style":856},[26,1056],{"className":1057,"style":860},[205],[26,1059,1061],{"className":1060},[210,864,865,213],[26,1062,251],{"className":1063},[47,48,213],[26,1065,221],{"className":1066},[220],[26,1068,1070],{"className":1069},[193],[26,1071,1073],{"className":1072,"style":878},[197],[26,1074],{},", ",[26,1077,1079],{"className":1078},[29],[26,1080,1082,1100],{"className":1081,"ariaHidden":34},[33],[26,1083,1085,1088,1091,1094,1097],{"className":1084},[38],[26,1086],{"className":1087,"style":781},[42],[26,1089,715],{"className":1090,"style":470},[47,179],[26,1092],{"className":1093,"style":358},[54],[26,1095,791],{"className":1096},[365],[26,1098],{"className":1099,"style":358},[54],[26,1101,1103,1106],{"className":1102},[38],[26,1104],{"className":1105,"style":801},[42],[26,1107,1109,1112],{"className":1108},[47],[26,1110,809],{"className":1111},[47,808],[26,1113,1115],{"className":1114},[184],[26,1116,1118],{"className":1117},[188],[26,1119,1121],{"className":1120},[193],[26,1122,1124],{"className":1123,"style":801},[197],[26,1125,1126,1129],{"style":824},[26,1127],{"className":1128,"style":206},[205],[26,1130,1132],{"className":1131},[210,211,212,213],[26,1133,1135,1187,1190],{"className":1134},[47,213],[26,1136,1138,1141],{"className":1137},[47,213],[26,1139,840],{"className":1140},[47,48,213],[26,1142,1144],{"className":1143},[184],[26,1145,1147,1179],{"className":1146},[188,189],[26,1148,1150,1176],{"className":1149},[193],[26,1151,1154],{"className":1152,"style":1153},[197],"height:0.2963em;",[26,1155,1156,1159],{"style":906},[26,1157],{"className":1158,"style":860},[205],[26,1160,1162],{"className":1161},[210,864,865,213],[26,1163,1165,1169,1173],{"className":1164},[47,213],[26,1166,1168],{"className":1167},[47,48,213],"o",[26,1170,1172],{"className":1171},[47,48,213],"u",[26,1174,158],{"className":1175},[47,48,213],[26,1177,221],{"className":1178},[220],[26,1180,1182],{"className":1181},[193],[26,1183,1185],{"className":1184,"style":929},[197],[26,1186],{},[26,1188,884],{"className":1189},[59,213],[26,1191,1193,1196],{"className":1192},[47,213],[26,1194,840],{"className":1195},[47,48,213],[26,1197,1199],{"className":1198},[184],[26,1200,1202,1222],{"className":1201},[188,189],[26,1203,1205,1219],{"className":1204},[193],[26,1206,1208],{"className":1207,"style":853},[197],[26,1209,1210,1213],{"style":856},[26,1211],{"className":1212,"style":860},[205],[26,1214,1216],{"className":1215},[210,864,865,213],[26,1217,251],{"className":1218},[47,48,213],[26,1220,221],{"className":1221},[220],[26,1223,1225],{"className":1224},[193],[26,1226,1228],{"className":1227,"style":878},[197],[26,1229],{},"\n— and, decisively, ",[21,1232,1233],{},"they are shared across every time step",". The same ",[26,1236,1238],{"className":1237},[29],[26,1239,1241],{"className":1240,"ariaHidden":34},[33],[26,1242,1244,1247],{"className":1243},[38],[26,1245],{"className":1246,"style":576},[42],[26,1248,407],{"className":1249},[47,179]," that\ncarries step ",[26,1252,1254],{"className":1253},[29],[26,1255,1257],{"className":1256,"ariaHidden":34},[33],[26,1258,1260,1263],{"className":1259},[38],[26,1261],{"className":1262,"style":70},[42],[26,1264,74],{"className":1265},[47],"'s context into step ",[26,1268,1270],{"className":1269},[29],[26,1271,1273],{"className":1272,"ariaHidden":34},[33],[26,1274,1276,1279],{"className":1275},[38],[26,1277],{"className":1278,"style":70},[42],[26,1280,1282],{"className":1281},[47],"2"," carries step ",[26,1285,1287],{"className":1286},[29],[26,1288,1290],{"className":1289,"ariaHidden":34},[33],[26,1291,1293,1296],{"className":1292},[38],[26,1294],{"className":1295,"style":70},[42],[26,1297,1299],{"className":1298},[47],"99","'s into step ",[26,1302,1304],{"className":1303},[29],[26,1305,1307],{"className":1306,"ariaHidden":34},[33],[26,1308,1310,1313],{"className":1309},[38],[26,1311],{"className":1312,"style":70},[42],[26,1314,1316],{"className":1315},[47],"100",". There is no\nper-position parameter and therefore no window; a sequence of any length is processed by the\nsame small set of weights.",[1319,1320],"tikz-figure",{"hash":1321},"96f30c51fe62e45e12efc9f46a062d97da62a477bbe34cd92ccd6ab67fce29c4",[11,1323,1324,1325,1328,1329,1332,1333,1385,1386,1389,1390,932,1405,1075,1420,1435],{},"Written this way the recurrence looks like new machinery, but it is not. If we ",[21,1326,1327],{},"unroll"," the\nnetwork in time — copy the cell once per time step and connect each copy's hidden state to the\nnext — the cycle disappears and what remains is an ordinary feedforward computation, just a\nvery deep one whose depth equals the sequence length. The copies are distinct in their\n",[94,1330,1331],{},"values"," (each ",[26,1334,1336],{"className":1335},[29],[26,1337,1339],{"className":1338,"ariaHidden":34},[33],[26,1340,1342,1345],{"className":1341},[38],[26,1343],{"className":1344,"style":314},[42],[26,1346,1348,1351],{"className":1347},[47],[26,1349,251],{"className":1350},[47,179],[26,1352,1354],{"className":1353},[184],[26,1355,1357,1377],{"className":1356},[188,189],[26,1358,1360,1374],{"className":1359},[193],[26,1361,1363],{"className":1362,"style":198},[197],[26,1364,1365,1368],{"style":201},[26,1366],{"className":1367,"style":206},[205],[26,1369,1371],{"className":1370},[210,211,212,213],[26,1372,158],{"className":1373},[47,48,213],[26,1375,221],{"className":1376},[220],[26,1378,1380],{"className":1379},[193],[26,1381,1383],{"className":1382,"style":228},[197],[26,1384],{}," differs) but identical in their ",[94,1387,1388],{},"weights"," (",[26,1391,1393],{"className":1392},[29],[26,1394,1396],{"className":1395,"ariaHidden":34},[33],[26,1397,1399,1402],{"className":1398},[38],[26,1400],{"className":1401,"style":576},[42],[26,1403,407],{"className":1404},[47,179],[26,1406,1408],{"className":1407},[29],[26,1409,1411],{"className":1410,"ariaHidden":34},[33],[26,1412,1414,1417],{"className":1413},[38],[26,1415],{"className":1416,"style":576},[42],[26,1418,715],{"className":1419,"style":470},[47,179],[26,1421,1423],{"className":1422},[29],[26,1424,1426],{"className":1425,"ariaHidden":34},[33],[26,1427,1429,1432],{"className":1428},[38],[26,1430],{"className":1431,"style":576},[42],[26,1433,471],{"className":1434,"style":470},[47,179]," are the same everywhere).",[1319,1437],{"hash":1438},"9bf56bc9c0cb8c460e6340888df2136a9d399c097ead1ae79b100ff814e7bd56",[1440,1441,1443],"callout",{"type":1442},"definition",[11,1444,1445,1448,1449,1563,1564,1750,1751,1075,1766,1781,1782,1797,1798,1850,1851,1961],{},[21,1446,1447],{},"Definition (Recurrent neural network)."," A network that computes a sequence of hidden\nstates ",[26,1450,1452],{"className":1451},[29],[26,1453,1455],{"className":1454,"ariaHidden":34},[33],[26,1456,1458,1462,1502,1505,1508,1512,1515,1518,1521],{"className":1457},[38],[26,1459],{"className":1460,"style":1461},[42],"height:0.8889em;vertical-align:-0.1944em;",[26,1463,1465,1468],{"className":1464},[47],[26,1466,251],{"className":1467},[47,179],[26,1469,1471],{"className":1470},[184],[26,1472,1474,1494],{"className":1473},[188,189],[26,1475,1477,1491],{"className":1476},[193],[26,1478,1480],{"className":1479,"style":264},[197],[26,1481,1482,1485],{"style":201},[26,1483],{"className":1484,"style":206},[205],[26,1486,1488],{"className":1487},[210,211,212,213],[26,1489,74],{"className":1490},[47,213],[26,1492,221],{"className":1493},[220],[26,1495,1497],{"className":1496},[193],[26,1498,1500],{"className":1499,"style":228},[197],[26,1501],{},[26,1503,524],{"className":1504},[523],[26,1506],{"className":1507,"style":392},[54],[26,1509,1511],{"className":1510},[396],"…",[26,1513],{"className":1514,"style":392},[54],[26,1516,524],{"className":1517},[523],[26,1519],{"className":1520,"style":392},[54],[26,1522,1524,1527],{"className":1523},[47],[26,1525,251],{"className":1526},[47,179],[26,1528,1530],{"className":1529},[184],[26,1531,1533,1555],{"className":1532},[188,189],[26,1534,1536,1552],{"className":1535},[193],[26,1537,1540],{"className":1538,"style":1539},[197],"height:0.1514em;",[26,1541,1542,1545],{"style":201},[26,1543],{"className":1544,"style":206},[205],[26,1546,1548],{"className":1547},[210,211,212,213],[26,1549,1551],{"className":1550},[47,48,213],"n",[26,1553,221],{"className":1554},[220],[26,1556,1558],{"className":1557},[193],[26,1559,1561],{"className":1560,"style":228},[197],[26,1562],{}," by the recurrence\n",[26,1565,1567],{"className":1566},[29],[26,1568,1570,1625,1698],{"className":1569,"ariaHidden":34},[33],[26,1571,1573,1576,1616,1619,1622],{"className":1572},[38],[26,1574],{"className":1575,"style":314},[42],[26,1577,1579,1582],{"className":1578},[47],[26,1580,251],{"className":1581},[47,179],[26,1583,1585],{"className":1584},[184],[26,1586,1588,1608],{"className":1587},[188,189],[26,1589,1591,1605],{"className":1590},[193],[26,1592,1594],{"className":1593,"style":198},[197],[26,1595,1596,1599],{"style":201},[26,1597],{"className":1598,"style":206},[205],[26,1600,1602],{"className":1601},[210,211,212,213],[26,1603,158],{"className":1604},[47,48,213],[26,1606,221],{"className":1607},[220],[26,1609,1611],{"className":1610},[193],[26,1612,1614],{"className":1613,"style":228},[197],[26,1615],{},[26,1617],{"className":1618,"style":358},[54],[26,1620,366],{"className":1621},[365],[26,1623],{"className":1624,"style":358},[54],[26,1626,1628,1631,1634,1637,1640,1689,1692,1695],{"className":1627},[38],[26,1629],{"className":1630,"style":379},[42],[26,1632,384],{"className":1633,"style":383},[47,48],[26,1635,403],{"className":1636},[400],[26,1638,407],{"className":1639},[47,179],[26,1641,1643,1646],{"className":1642},[47],[26,1644,251],{"className":1645},[47,179],[26,1647,1649],{"className":1648},[184],[26,1650,1652,1681],{"className":1651},[188,189],[26,1653,1655,1678],{"className":1654},[193],[26,1656,1658],{"className":1657,"style":264},[197],[26,1659,1660,1663],{"style":201},[26,1661],{"className":1662,"style":206},[205],[26,1664,1666],{"className":1665},[210,211,212,213],[26,1667,1669,1672,1675],{"className":1668},[47,213],[26,1670,158],{"className":1671},[47,48,213],[26,1673,60],{"className":1674},[59,213],[26,1676,74],{"className":1677},[47,213],[26,1679,221],{"className":1680},[220],[26,1682,1684],{"className":1683},[193],[26,1685,1687],{"className":1686,"style":294},[197],[26,1688],{},[26,1690],{"className":1691,"style":55},[54],[26,1693,463],{"className":1694},[59],[26,1696],{"className":1697,"style":55},[54],[26,1699,1701,1704,1707,1747],{"className":1700},[38],[26,1702],{"className":1703,"style":379},[42],[26,1705,471],{"className":1706,"style":470},[47,179],[26,1708,1710,1713],{"className":1709},[47],[26,1711,180],{"className":1712},[47,179],[26,1714,1716],{"className":1715},[184],[26,1717,1719,1739],{"className":1718},[188,189],[26,1720,1722,1736],{"className":1721},[193],[26,1723,1725],{"className":1724,"style":198},[197],[26,1726,1727,1730],{"style":201},[26,1728],{"className":1729,"style":206},[205],[26,1731,1733],{"className":1732},[210,211,212,213],[26,1734,158],{"className":1735},[47,48,213],[26,1737,221],{"className":1738},[220],[26,1740,1742],{"className":1741},[193],[26,1743,1745],{"className":1744,"style":228},[197],[26,1746],{},[26,1748,516],{"className":1749},[515],", sharing the\nweights ",[26,1752,1754],{"className":1753},[29],[26,1755,1757],{"className":1756,"ariaHidden":34},[33],[26,1758,1760,1763],{"className":1759},[38],[26,1761],{"className":1762,"style":576},[42],[26,1764,407],{"className":1765},[47,179],[26,1767,1769],{"className":1768},[29],[26,1770,1772],{"className":1771,"ariaHidden":34},[33],[26,1773,1775,1778],{"className":1774},[38],[26,1776],{"className":1777,"style":576},[42],[26,1779,471],{"className":1780,"style":470},[47,179]," across all ",[26,1783,1785],{"className":1784},[29],[26,1786,1788],{"className":1787,"ariaHidden":34},[33],[26,1789,1791,1794],{"className":1790},[38],[26,1792],{"className":1793,"style":154},[42],[26,1795,158],{"className":1796},[47,48],". The state ",[26,1799,1801],{"className":1800},[29],[26,1802,1804],{"className":1803,"ariaHidden":34},[33],[26,1805,1807,1810],{"className":1806},[38],[26,1808],{"className":1809,"style":314},[42],[26,1811,1813,1816],{"className":1812},[47],[26,1814,251],{"className":1815},[47,179],[26,1817,1819],{"className":1818},[184],[26,1820,1822,1842],{"className":1821},[188,189],[26,1823,1825,1839],{"className":1824},[193],[26,1826,1828],{"className":1827,"style":198},[197],[26,1829,1830,1833],{"style":201},[26,1831],{"className":1832,"style":206},[205],[26,1834,1836],{"className":1835},[210,211,212,213],[26,1837,158],{"className":1838},[47,48,213],[26,1840,221],{"className":1841},[220],[26,1843,1845],{"className":1844},[193],[26,1846,1848],{"className":1847,"style":228},[197],[26,1849],{}," encodes the\nwhole prefix ",[26,1852,1854],{"className":1853},[29],[26,1855,1857],{"className":1856,"ariaHidden":34},[33],[26,1858,1860,1863,1903,1906,1909,1912,1915,1918,1921],{"className":1859},[38],[26,1861],{"className":1862,"style":628},[42],[26,1864,1866,1869],{"className":1865},[47],[26,1867,180],{"className":1868},[47,179],[26,1870,1872],{"className":1871},[184],[26,1873,1875,1895],{"className":1874},[188,189],[26,1876,1878,1892],{"className":1877},[193],[26,1879,1881],{"className":1880,"style":264},[197],[26,1882,1883,1886],{"style":201},[26,1884],{"className":1885,"style":206},[205],[26,1887,1889],{"className":1888},[210,211,212,213],[26,1890,74],{"className":1891},[47,213],[26,1893,221],{"className":1894},[220],[26,1896,1898],{"className":1897},[193],[26,1899,1901],{"className":1900,"style":228},[197],[26,1902],{},[26,1904,524],{"className":1905},[523],[26,1907],{"className":1908,"style":392},[54],[26,1910,1511],{"className":1911},[396],[26,1913],{"className":1914,"style":392},[54],[26,1916,524],{"className":1917},[523],[26,1919],{"className":1920,"style":392},[54],[26,1922,1924,1927],{"className":1923},[47],[26,1925,180],{"className":1926},[47,179],[26,1928,1930],{"className":1929},[184],[26,1931,1933,1953],{"className":1932},[188,189],[26,1934,1936,1950],{"className":1935},[193],[26,1937,1939],{"className":1938,"style":198},[197],[26,1940,1941,1944],{"style":201},[26,1942],{"className":1943,"style":206},[205],[26,1945,1947],{"className":1946},[210,211,212,213],[26,1948,158],{"className":1949},[47,48,213],[26,1951,221],{"className":1952},[220],[26,1954,1956],{"className":1955},[193],[26,1957,1959],{"className":1958,"style":228},[197],[26,1960],{},", so context is not bounded by a fixed\nwindow.",[1963,1964,1966],"h3",{"id":1965},"forward-inference","Forward inference",[11,1968,1969,1970,2022,2023,2084],{},"Because ",[26,1971,1973],{"className":1972},[29],[26,1974,1976],{"className":1975,"ariaHidden":34},[33],[26,1977,1979,1982],{"className":1978},[38],[26,1980],{"className":1981,"style":314},[42],[26,1983,1985,1988],{"className":1984},[47],[26,1986,251],{"className":1987},[47,179],[26,1989,1991],{"className":1990},[184],[26,1992,1994,2014],{"className":1993},[188,189],[26,1995,1997,2011],{"className":1996},[193],[26,1998,2000],{"className":1999,"style":198},[197],[26,2001,2002,2005],{"style":201},[26,2003],{"className":2004,"style":206},[205],[26,2006,2008],{"className":2007},[210,211,212,213],[26,2009,158],{"className":2010},[47,48,213],[26,2012,221],{"className":2013},[220],[26,2015,2017],{"className":2016},[193],[26,2018,2020],{"className":2019,"style":228},[197],[26,2021],{}," needs ",[26,2024,2026],{"className":2025},[29],[26,2027,2029],{"className":2028,"ariaHidden":34},[33],[26,2030,2032,2035],{"className":2031},[38],[26,2033],{"className":2034,"style":244},[42],[26,2036,2038,2041],{"className":2037},[47],[26,2039,251],{"className":2040},[47,179],[26,2042,2044],{"className":2043},[184],[26,2045,2047,2076],{"className":2046},[188,189],[26,2048,2050,2073],{"className":2049},[193],[26,2051,2053],{"className":2052,"style":264},[197],[26,2054,2055,2058],{"style":201},[26,2056],{"className":2057,"style":206},[205],[26,2059,2061],{"className":2060},[210,211,212,213],[26,2062,2064,2067,2070],{"className":2063},[47,213],[26,2065,158],{"className":2066},[47,48,213],[26,2068,60],{"className":2069},[59,213],[26,2071,74],{"className":2072},[47,213],[26,2074,221],{"className":2075},[220],[26,2077,2079],{"className":2078},[193],[26,2080,2082],{"className":2081,"style":294},[197],[26,2083],{},", inference is inherently sequential: we sweep\nleft to right, and at each step compute the hidden state from the previous one and read off\nthe output. The whole forward pass is four lines.",[2086,2087,2091],"pre",{"className":2088,"code":2089,"language":2090,"meta":6,"style":6},"language-algorithm shiki shiki-themes Vesper Light - Orange Boost (Quick Open Adjusted) vesper","caption: $\\textsc{Forward-RNN}(\\mathbf{x}, \\mathit{network})$ — map an input sequence to outputs\ninput: sequence $\\mathbf{x} = \\mathbf{x}_1, \\ldots, \\mathbf{x}_n$; weights $\\mathbf{U}, \\mathbf{V}, \\mathbf{W}$\n$\\mathbf{h}_0 \\gets \\mathbf{0}$\nfor $t = 1$ to $n$ do\n  $\\mathbf{h}_t \\gets g(\\mathbf{U}\\mathbf{h}_{t-1} + \\mathbf{W}\\mathbf{x}_t)$ \u002F\u002F carry context forward\n  $\\mathbf{y}_t \\gets f(\\mathbf{V}\\mathbf{h}_t)$ \u002F\u002F read the output\nreturn $\\mathbf{y}_1, \\ldots, \\mathbf{y}_n$\n","algorithm",[2092,2093,2094,2101,2107,2113,2119,2125,2131],"code",{"__ignoreMap":6},[26,2095,2098],{"class":2096,"line":2097},"line",1,[26,2099,2100],{},"caption: $\\textsc{Forward-RNN}(\\mathbf{x}, \\mathit{network})$ — map an input sequence to outputs\n",[26,2102,2104],{"class":2096,"line":2103},2,[26,2105,2106],{},"input: sequence $\\mathbf{x} = \\mathbf{x}_1, \\ldots, \\mathbf{x}_n$; weights $\\mathbf{U}, \\mathbf{V}, \\mathbf{W}$\n",[26,2108,2110],{"class":2096,"line":2109},3,[26,2111,2112],{},"$\\mathbf{h}_0 \\gets \\mathbf{0}$\n",[26,2114,2116],{"class":2096,"line":2115},4,[26,2117,2118],{},"for $t = 1$ to $n$ do\n",[26,2120,2122],{"class":2096,"line":2121},5,[26,2123,2124],{},"  $\\mathbf{h}_t \\gets g(\\mathbf{U}\\mathbf{h}_{t-1} + \\mathbf{W}\\mathbf{x}_t)$ \u002F\u002F carry context forward\n",[26,2126,2128],{"class":2096,"line":2127},6,[26,2129,2130],{},"  $\\mathbf{y}_t \\gets f(\\mathbf{V}\\mathbf{h}_t)$ \u002F\u002F read the output\n",[26,2132,2134],{"class":2096,"line":2133},7,[26,2135,2136],{},"return $\\mathbf{y}_1, \\ldots, \\mathbf{y}_n$\n",[11,2138,2139],{},"The sequential dependence is both the strength (unbounded context) and the weakness (no\nparallelism across time) of the architecture, and that lack of parallelism is what later\nmotivates the transformer.",[1963,2141,2143],{"id":2142},"worked-example-a-two-step-forward-pass","Worked example: a two-step forward pass",[11,2145,2146,2147,2180,2181,1075,2215,2249,2250,2321,2322,2394,2395,765],{},"Run the recurrence with tiny numbers. Take one-dimensional hidden and input states,\n",[26,2148,2150],{"className":2149},[29],[26,2151,2153,2171],{"className":2152,"ariaHidden":34},[33],[26,2154,2156,2159,2162,2165,2168],{"className":2155},[38],[26,2157],{"className":2158,"style":540},[42],[26,2160,384],{"className":2161,"style":383},[47,48],[26,2163],{"className":2164,"style":358},[54],[26,2166,366],{"className":2167},[365],[26,2169],{"className":2170,"style":358},[54],[26,2172,2174,2177],{"className":2173},[38],[26,2175],{"className":2176,"style":557},[42],[26,2178,562],{"className":2179},[561],", and weights ",[26,2182,2184],{"className":2183},[29],[26,2185,2187,2205],{"className":2186,"ariaHidden":34},[33],[26,2188,2190,2193,2196,2199,2202],{"className":2189},[38],[26,2191],{"className":2192,"style":576},[42],[26,2194,471],{"className":2195,"style":470},[47,179],[26,2197],{"className":2198,"style":358},[54],[26,2200,366],{"className":2201},[365],[26,2203],{"className":2204,"style":358},[54],[26,2206,2208,2211],{"className":2207},[38],[26,2209],{"className":2210,"style":70},[42],[26,2212,2214],{"className":2213},[47],"0.8",[26,2216,2218],{"className":2217},[29],[26,2219,2221,2239],{"className":2220,"ariaHidden":34},[33],[26,2222,2224,2227,2230,2233,2236],{"className":2223},[38],[26,2225],{"className":2226,"style":576},[42],[26,2228,407],{"className":2229},[47,179],[26,2231],{"className":2232,"style":358},[54],[26,2234,366],{"className":2235},[365],[26,2237],{"className":2238,"style":358},[54],[26,2240,2242,2245],{"className":2241},[38],[26,2243],{"className":2244,"style":70},[42],[26,2246,2248],{"className":2247},[47],"0.5",", with ",[26,2251,2253],{"className":2252},[29],[26,2254,2256,2312],{"className":2255,"ariaHidden":34},[33],[26,2257,2259,2262,2303,2306,2309],{"className":2258},[38],[26,2260],{"className":2261,"style":314},[42],[26,2263,2265,2268],{"className":2264},[47],[26,2266,251],{"className":2267},[47,48],[26,2269,2271],{"className":2270},[184],[26,2272,2274,2295],{"className":2273},[188,189],[26,2275,2277,2292],{"className":2276},[193],[26,2278,2280],{"className":2279,"style":264},[197],[26,2281,2282,2285],{"style":201},[26,2283],{"className":2284,"style":206},[205],[26,2286,2288],{"className":2287},[210,211,212,213],[26,2289,2291],{"className":2290},[47,213],"0",[26,2293,221],{"className":2294},[220],[26,2296,2298],{"className":2297},[193],[26,2299,2301],{"className":2300,"style":228},[197],[26,2302],{},[26,2304],{"className":2305,"style":358},[54],[26,2307,366],{"className":2308},[365],[26,2310],{"className":2311,"style":358},[54],[26,2313,2315,2318],{"className":2314},[38],[26,2316],{"className":2317,"style":70},[42],[26,2319,2291],{"className":2320},[47],". Feed\nthe inputs ",[26,2323,2325],{"className":2324},[29],[26,2326,2328,2384],{"className":2327,"ariaHidden":34},[33],[26,2329,2331,2335,2375,2378,2381],{"className":2330},[38],[26,2332],{"className":2333,"style":2334},[42],"height:0.5806em;vertical-align:-0.15em;",[26,2336,2338,2341],{"className":2337},[47],[26,2339,180],{"className":2340},[47,48],[26,2342,2344],{"className":2343},[184],[26,2345,2347,2367],{"className":2346},[188,189],[26,2348,2350,2364],{"className":2349},[193],[26,2351,2353],{"className":2352,"style":264},[197],[26,2354,2355,2358],{"style":201},[26,2356],{"className":2357,"style":206},[205],[26,2359,2361],{"className":2360},[210,211,212,213],[26,2362,74],{"className":2363},[47,213],[26,2365,221],{"className":2366},[220],[26,2368,2370],{"className":2369},[193],[26,2371,2373],{"className":2372,"style":228},[197],[26,2374],{},[26,2376],{"className":2377,"style":358},[54],[26,2379,366],{"className":2380},[365],[26,2382],{"className":2383,"style":358},[54],[26,2385,2387,2390],{"className":2386},[38],[26,2388],{"className":2389,"style":70},[42],[26,2391,2393],{"className":2392},[47],"1.0"," then ",[26,2396,2398],{"className":2397},[29],[26,2399,2401,2456],{"className":2400,"ariaHidden":34},[33],[26,2402,2404,2407,2447,2450,2453],{"className":2403},[38],[26,2405],{"className":2406,"style":2334},[42],[26,2408,2410,2413],{"className":2409},[47],[26,2411,180],{"className":2412},[47,48],[26,2414,2416],{"className":2415},[184],[26,2417,2419,2439],{"className":2418},[188,189],[26,2420,2422,2436],{"className":2421},[193],[26,2423,2425],{"className":2424,"style":264},[197],[26,2426,2427,2430],{"style":201},[26,2428],{"className":2429,"style":206},[205],[26,2431,2433],{"className":2432},[210,211,212,213],[26,2434,1282],{"className":2435},[47,213],[26,2437,221],{"className":2438},[220],[26,2440,2442],{"className":2441},[193],[26,2443,2445],{"className":2444,"style":228},[197],[26,2446],{},[26,2448],{"className":2449,"style":358},[54],[26,2451,366],{"className":2452},[365],[26,2454],{"className":2455,"style":358},[54],[26,2457,2459,2463,2466],{"className":2458},[38],[26,2460],{"className":2461,"style":2462},[42],"height:0.7278em;vertical-align:-0.0833em;",[26,2464,60],{"className":2465},[47],[26,2467,2248],{"className":2468},[47],[11,2470,2471,2474,2475,2661,2662,2760,2761,765],{},[21,2472,2473],{},"Step 1."," The pre-activation is ",[26,2476,2478],{"className":2477},[29],[26,2479,2481,2539,2598,2625,2652],{"className":2480,"ariaHidden":34},[33],[26,2482,2484,2487,2490,2530,2533,2536],{"className":2483},[38],[26,2485],{"className":2486,"style":314},[42],[26,2488,407],{"className":2489},[47,179],[26,2491,2493,2496],{"className":2492},[47],[26,2494,251],{"className":2495},[47,48],[26,2497,2499],{"className":2498},[184],[26,2500,2502,2522],{"className":2501},[188,189],[26,2503,2505,2519],{"className":2504},[193],[26,2506,2508],{"className":2507,"style":264},[197],[26,2509,2510,2513],{"style":201},[26,2511],{"className":2512,"style":206},[205],[26,2514,2516],{"className":2515},[210,211,212,213],[26,2517,2291],{"className":2518},[47,213],[26,2520,221],{"className":2521},[220],[26,2523,2525],{"className":2524},[193],[26,2526,2528],{"className":2527,"style":228},[197],[26,2529],{},[26,2531],{"className":2532,"style":55},[54],[26,2534,463],{"className":2535},[59],[26,2537],{"className":2538,"style":55},[54],[26,2540,2542,2546,2549,2589,2592,2595],{"className":2541},[38],[26,2543],{"className":2544,"style":2545},[42],"height:0.8361em;vertical-align:-0.15em;",[26,2547,471],{"className":2548,"style":470},[47,179],[26,2550,2552,2555],{"className":2551},[47],[26,2553,180],{"className":2554},[47,48],[26,2556,2558],{"className":2557},[184],[26,2559,2561,2581],{"className":2560},[188,189],[26,2562,2564,2578],{"className":2563},[193],[26,2565,2567],{"className":2566,"style":264},[197],[26,2568,2569,2572],{"style":201},[26,2570],{"className":2571,"style":206},[205],[26,2573,2575],{"className":2574},[210,211,212,213],[26,2576,74],{"className":2577},[47,213],[26,2579,221],{"className":2580},[220],[26,2582,2584],{"className":2583},[193],[26,2585,2587],{"className":2586,"style":228},[197],[26,2588],{},[26,2590],{"className":2591,"style":358},[54],[26,2593,366],{"className":2594},[365],[26,2596],{"className":2597,"style":358},[54],[26,2599,2601,2604,2607,2610,2613,2616,2619,2622],{"className":2600},[38],[26,2602],{"className":2603,"style":379},[42],[26,2605,2248],{"className":2606},[47],[26,2608,403],{"className":2609},[400],[26,2611,2291],{"className":2612},[47],[26,2614,516],{"className":2615},[515],[26,2617],{"className":2618,"style":55},[54],[26,2620,463],{"className":2621},[59],[26,2623],{"className":2624,"style":55},[54],[26,2626,2628,2631,2634,2637,2640,2643,2646,2649],{"className":2627},[38],[26,2629],{"className":2630,"style":379},[42],[26,2632,2214],{"className":2633},[47],[26,2635,403],{"className":2636},[400],[26,2638,2393],{"className":2639},[47],[26,2641,516],{"className":2642},[515],[26,2644],{"className":2645,"style":358},[54],[26,2647,366],{"className":2648},[365],[26,2650],{"className":2651,"style":358},[54],[26,2653,2655,2658],{"className":2654},[38],[26,2656],{"className":2657,"style":70},[42],[26,2659,2214],{"className":2660},[47],", so ",[26,2663,2665],{"className":2664},[29],[26,2666,2668,2723,2750],{"className":2667,"ariaHidden":34},[33],[26,2669,2671,2674,2714,2717,2720],{"className":2670},[38],[26,2672],{"className":2673,"style":314},[42],[26,2675,2677,2680],{"className":2676},[47],[26,2678,251],{"className":2679},[47,48],[26,2681,2683],{"className":2682},[184],[26,2684,2686,2706],{"className":2685},[188,189],[26,2687,2689,2703],{"className":2688},[193],[26,2690,2692],{"className":2691,"style":264},[197],[26,2693,2694,2697],{"style":201},[26,2695],{"className":2696,"style":206},[205],[26,2698,2700],{"className":2699},[210,211,212,213],[26,2701,74],{"className":2702},[47,213],[26,2704,221],{"className":2705},[220],[26,2707,2709],{"className":2708},[193],[26,2710,2712],{"className":2711,"style":228},[197],[26,2713],{},[26,2715],{"className":2716,"style":358},[54],[26,2718,366],{"className":2719},[365],[26,2721],{"className":2722,"style":358},[54],[26,2724,2726,2729,2732,2735,2738,2741,2744,2747],{"className":2725},[38],[26,2727],{"className":2728,"style":379},[42],[26,2730,562],{"className":2731},[561],[26,2733,403],{"className":2734},[400],[26,2736,2214],{"className":2737},[47],[26,2739,516],{"className":2740},[515],[26,2742],{"className":2743,"style":358},[54],[26,2745,366],{"className":2746},[365],[26,2748],{"className":2749,"style":358},[54],[26,2751,2753,2756],{"className":2752},[38],[26,2754],{"className":2755,"style":70},[42],[26,2757,2759],{"className":2758},[47],"0.664",". The state now carries a trace\nof ",[26,2762,2764],{"className":2763},[29],[26,2765,2767],{"className":2766,"ariaHidden":34},[33],[26,2768,2770,2773],{"className":2769},[38],[26,2771],{"className":2772,"style":2334},[42],[26,2774,2776,2779],{"className":2775},[47],[26,2777,180],{"className":2778},[47,48],[26,2780,2782],{"className":2781},[184],[26,2783,2785,2805],{"className":2784},[188,189],[26,2786,2788,2802],{"className":2787},[193],[26,2789,2791],{"className":2790,"style":264},[197],[26,2792,2793,2796],{"style":201},[26,2794],{"className":2795,"style":206},[205],[26,2797,2799],{"className":2798},[210,211,212,213],[26,2800,74],{"className":2801},[47,213],[26,2803,221],{"className":2804},[220],[26,2806,2808],{"className":2807},[193],[26,2809,2811],{"className":2810,"style":228},[197],[26,2812],{},[11,2814,2815,2474,2818,2661,3048,3151,3152,3204,3205,3257,3258,3257,3261,3313,3314,3369,3370,3385],{},[21,2816,2817],{},"Step 2.",[26,2819,2821],{"className":2820},[29],[26,2822,2824,2882,2940,2967,2997,3016,3035],{"className":2823,"ariaHidden":34},[33],[26,2825,2827,2830,2833,2873,2876,2879],{"className":2826},[38],[26,2828],{"className":2829,"style":314},[42],[26,2831,407],{"className":2832},[47,179],[26,2834,2836,2839],{"className":2835},[47],[26,2837,251],{"className":2838},[47,48],[26,2840,2842],{"className":2841},[184],[26,2843,2845,2865],{"className":2844},[188,189],[26,2846,2848,2862],{"className":2847},[193],[26,2849,2851],{"className":2850,"style":264},[197],[26,2852,2853,2856],{"style":201},[26,2854],{"className":2855,"style":206},[205],[26,2857,2859],{"className":2858},[210,211,212,213],[26,2860,74],{"className":2861},[47,213],[26,2863,221],{"className":2864},[220],[26,2866,2868],{"className":2867},[193],[26,2869,2871],{"className":2870,"style":228},[197],[26,2872],{},[26,2874],{"className":2875,"style":55},[54],[26,2877,463],{"className":2878},[59],[26,2880],{"className":2881,"style":55},[54],[26,2883,2885,2888,2891,2931,2934,2937],{"className":2884},[38],[26,2886],{"className":2887,"style":2545},[42],[26,2889,471],{"className":2890,"style":470},[47,179],[26,2892,2894,2897],{"className":2893},[47],[26,2895,180],{"className":2896},[47,48],[26,2898,2900],{"className":2899},[184],[26,2901,2903,2923],{"className":2902},[188,189],[26,2904,2906,2920],{"className":2905},[193],[26,2907,2909],{"className":2908,"style":264},[197],[26,2910,2911,2914],{"style":201},[26,2912],{"className":2913,"style":206},[205],[26,2915,2917],{"className":2916},[210,211,212,213],[26,2918,1282],{"className":2919},[47,213],[26,2921,221],{"className":2922},[220],[26,2924,2926],{"className":2925},[193],[26,2927,2929],{"className":2928,"style":228},[197],[26,2930],{},[26,2932],{"className":2933,"style":358},[54],[26,2935,366],{"className":2936},[365],[26,2938],{"className":2939,"style":358},[54],[26,2941,2943,2946,2949,2952,2955,2958,2961,2964],{"className":2942},[38],[26,2944],{"className":2945,"style":379},[42],[26,2947,2248],{"className":2948},[47],[26,2950,403],{"className":2951},[400],[26,2953,2759],{"className":2954},[47],[26,2956,516],{"className":2957},[515],[26,2959],{"className":2960,"style":55},[54],[26,2962,463],{"className":2963},[59],[26,2965],{"className":2966,"style":55},[54],[26,2968,2970,2973,2976,2979,2982,2985,2988,2991,2994],{"className":2969},[38],[26,2971],{"className":2972,"style":379},[42],[26,2974,2214],{"className":2975},[47],[26,2977,403],{"className":2978},[400],[26,2980,60],{"className":2981},[47],[26,2983,2248],{"className":2984},[47],[26,2986,516],{"className":2987},[515],[26,2989],{"className":2990,"style":358},[54],[26,2992,366],{"className":2993},[365],[26,2995],{"className":2996,"style":358},[54],[26,2998,3000,3003,3007,3010,3013],{"className":2999},[38],[26,3001],{"className":3002,"style":2462},[42],[26,3004,3006],{"className":3005},[47],"0.332",[26,3008],{"className":3009,"style":55},[54],[26,3011,60],{"className":3012},[59],[26,3014],{"className":3015,"style":55},[54],[26,3017,3019,3022,3026,3029,3032],{"className":3018},[38],[26,3020],{"className":3021,"style":70},[42],[26,3023,3025],{"className":3024},[47],"0.400",[26,3027],{"className":3028,"style":358},[54],[26,3030,366],{"className":3031},[365],[26,3033],{"className":3034,"style":358},[54],[26,3036,3038,3041,3044],{"className":3037},[38],[26,3039],{"className":3040,"style":2462},[42],[26,3042,60],{"className":3043},[47],[26,3045,3047],{"className":3046},[47],"0.068",[26,3049,3051],{"className":3050},[29],[26,3052,3054,3109,3139],{"className":3053,"ariaHidden":34},[33],[26,3055,3057,3060,3100,3103,3106],{"className":3056},[38],[26,3058],{"className":3059,"style":314},[42],[26,3061,3063,3066],{"className":3062},[47],[26,3064,251],{"className":3065},[47,48],[26,3067,3069],{"className":3068},[184],[26,3070,3072,3092],{"className":3071},[188,189],[26,3073,3075,3089],{"className":3074},[193],[26,3076,3078],{"className":3077,"style":264},[197],[26,3079,3080,3083],{"style":201},[26,3081],{"className":3082,"style":206},[205],[26,3084,3086],{"className":3085},[210,211,212,213],[26,3087,1282],{"className":3088},[47,213],[26,3090,221],{"className":3091},[220],[26,3093,3095],{"className":3094},[193],[26,3096,3098],{"className":3097,"style":228},[197],[26,3099],{},[26,3101],{"className":3102,"style":358},[54],[26,3104,366],{"className":3105},[365],[26,3107],{"className":3108,"style":358},[54],[26,3110,3112,3115,3118,3121,3124,3127,3130,3133,3136],{"className":3111},[38],[26,3113],{"className":3114,"style":379},[42],[26,3116,562],{"className":3117},[561],[26,3119,403],{"className":3120},[400],[26,3122,60],{"className":3123},[47],[26,3125,3047],{"className":3126},[47],[26,3128,516],{"className":3129},[515],[26,3131],{"className":3132,"style":358},[54],[26,3134,366],{"className":3135},[365],[26,3137],{"className":3138,"style":358},[54],[26,3140,3142,3145,3148],{"className":3141},[38],[26,3143],{"className":3144,"style":2462},[42],[26,3146,60],{"className":3147},[47],[26,3149,3047],{"className":3150},[47],". Notice that ",[26,3153,3155],{"className":3154},[29],[26,3156,3158],{"className":3157,"ariaHidden":34},[33],[26,3159,3161,3164],{"className":3160},[38],[26,3162],{"className":3163,"style":314},[42],[26,3165,3167,3170],{"className":3166},[47],[26,3168,251],{"className":3169},[47,48],[26,3171,3173],{"className":3172},[184],[26,3174,3176,3196],{"className":3175},[188,189],[26,3177,3179,3193],{"className":3178},[193],[26,3180,3182],{"className":3181,"style":264},[197],[26,3183,3184,3187],{"style":201},[26,3185],{"className":3186,"style":206},[205],[26,3188,3190],{"className":3189},[210,211,212,213],[26,3191,1282],{"className":3192},[47,213],[26,3194,221],{"className":3195},[220],[26,3197,3199],{"className":3198},[193],[26,3200,3202],{"className":3201,"style":228},[197],[26,3203],{}," depends\non ",[26,3206,3208],{"className":3207},[29],[26,3209,3211],{"className":3210,"ariaHidden":34},[33],[26,3212,3214,3217],{"className":3213},[38],[26,3215],{"className":3216,"style":2334},[42],[26,3218,3220,3223],{"className":3219},[47],[26,3221,180],{"className":3222},[47,48],[26,3224,3226],{"className":3225},[184],[26,3227,3229,3249],{"className":3228},[188,189],[26,3230,3232,3246],{"className":3231},[193],[26,3233,3235],{"className":3234,"style":264},[197],[26,3236,3237,3240],{"style":201},[26,3238],{"className":3239,"style":206},[205],[26,3241,3243],{"className":3242},[210,211,212,213],[26,3244,74],{"className":3245},[47,213],[26,3247,221],{"className":3248},[220],[26,3250,3252],{"className":3251},[193],[26,3253,3255],{"className":3254,"style":228},[197],[26,3256],{}," ",[94,3259,3260],{},"only through",[26,3262,3264],{"className":3263},[29],[26,3265,3267],{"className":3266,"ariaHidden":34},[33],[26,3268,3270,3273],{"className":3269},[38],[26,3271],{"className":3272,"style":314},[42],[26,3274,3276,3279],{"className":3275},[47],[26,3277,251],{"className":3278},[47,48],[26,3280,3282],{"className":3281},[184],[26,3283,3285,3305],{"className":3284},[188,189],[26,3286,3288,3302],{"className":3287},[193],[26,3289,3291],{"className":3290,"style":264},[197],[26,3292,3293,3296],{"style":201},[26,3294],{"className":3295,"style":206},[205],[26,3297,3299],{"className":3298},[210,211,212,213],[26,3300,74],{"className":3301},[47,213],[26,3303,221],{"className":3304},[220],[26,3306,3308],{"className":3307},[193],[26,3309,3311],{"className":3310,"style":228},[197],[26,3312],{}," — the recurrent term ",[26,3315,3317],{"className":3316},[29],[26,3318,3320],{"className":3319,"ariaHidden":34},[33],[26,3321,3323,3326,3329],{"className":3322},[38],[26,3324],{"className":3325,"style":314},[42],[26,3327,407],{"className":3328},[47,179],[26,3330,3332,3335],{"className":3331},[47],[26,3333,251],{"className":3334},[47,48],[26,3336,3338],{"className":3337},[184],[26,3339,3341,3361],{"className":3340},[188,189],[26,3342,3344,3358],{"className":3343},[193],[26,3345,3347],{"className":3346,"style":264},[197],[26,3348,3349,3352],{"style":201},[26,3350],{"className":3351,"style":206},[205],[26,3353,3355],{"className":3354},[210,211,212,213],[26,3356,74],{"className":3357},[47,213],[26,3359,221],{"className":3360},[220],[26,3362,3364],{"className":3363},[193],[26,3365,3367],{"className":3366,"style":228},[197],[26,3368],{}," is the single channel\nby which step 1's information reaches step 2. This is the whole mechanism: everything the\npast contributes to the present flows through the one hidden wire, which is also why a\nweak recurrent weight (here ",[26,3371,3373],{"className":3372},[29],[26,3374,3376],{"className":3375,"ariaHidden":34},[33],[26,3377,3379,3382],{"className":3378},[38],[26,3380],{"className":3381,"style":70},[42],[26,3383,2248],{"className":3384},[47],") attenuates old context, foreshadowing the vanishing\ngradient.",[127,3387,3389],{"id":3388},"rnns-as-language-models","RNNs as language models",[11,3391,3392,3393,3400,3401,3453],{},"To use an RNN as a language model, run it over a corpus and, at each step, have it predict the\nnext word from the current word and the hidden state.",[117,3394,3395],{},[15,3396,1282],{"href":3397,"ariaDescribedBy":3398,"dataFootnoteRef":6,"id":3399},"#user-content-fn-jm-lm",[123],"user-content-fnref-jm-lm"," Unlike the fixed-window neural\nmodel, the RNN has no context limit: ",[26,3402,3404],{"className":3403},[29],[26,3405,3407],{"className":3406,"ariaHidden":34},[33],[26,3408,3410,3413],{"className":3409},[38],[26,3411],{"className":3412,"style":314},[42],[26,3414,3416,3419],{"className":3415},[47],[26,3417,251],{"className":3418},[47,179],[26,3420,3422],{"className":3421},[184],[26,3423,3425,3445],{"className":3424},[188,189],[26,3426,3428,3442],{"className":3427},[193],[26,3429,3431],{"className":3430,"style":198},[197],[26,3432,3433,3436],{"style":201},[26,3434],{"className":3435,"style":206},[205],[26,3437,3439],{"className":3438},[210,211,212,213],[26,3440,158],{"className":3441},[47,48,213],[26,3443,221],{"className":3444},[220],[26,3446,3448],{"className":3447},[193],[26,3449,3451],{"className":3450,"style":228},[197],[26,3452],{}," can in principle carry information from the\nvery first word of the sequence.",[11,3455,3456,3457,3472,3473,3489],{},"At step ",[26,3458,3460],{"className":3459},[29],[26,3461,3463],{"className":3462,"ariaHidden":34},[33],[26,3464,3466,3469],{"className":3465},[38],[26,3467],{"className":3468,"style":154},[42],[26,3470,158],{"className":3471},[47,48]," the current word is looked up in the embedding matrix ",[26,3474,3476],{"className":3475},[29],[26,3477,3479],{"className":3478,"ariaHidden":34},[33],[26,3480,3482,3485],{"className":3481},[38],[26,3483],{"className":3484,"style":576},[42],[26,3486,3488],{"className":3487},[47,179],"E",", combined with\nthe previous hidden state, and turned into a distribution over the whole vocabulary:",[26,3491,3493],{"className":3492},[301],[26,3494,3496],{"className":3495},[29],[26,3497,3499,3561,3675,3870],{"className":3498,"ariaHidden":34},[33],[26,3500,3502,3505,3546,3549,3552,3555,3558],{"className":3501},[38],[26,3503],{"className":3504,"style":172},[42],[26,3506,3508,3512],{"className":3507},[47],[26,3509,3511],{"className":3510},[47,179],"e",[26,3513,3515],{"className":3514},[184],[26,3516,3518,3538],{"className":3517},[188,189],[26,3519,3521,3535],{"className":3520},[193],[26,3522,3524],{"className":3523,"style":198},[197],[26,3525,3526,3529],{"style":201},[26,3527],{"className":3528,"style":206},[205],[26,3530,3532],{"className":3531},[210,211,212,213],[26,3533,158],{"className":3534},[47,48,213],[26,3536,221],{"className":3537},[220],[26,3539,3541],{"className":3540},[193],[26,3542,3544],{"className":3543,"style":228},[197],[26,3545],{},[26,3547],{"className":3548,"style":358},[54],[26,3550],{"className":3551,"style":358},[54],[26,3553,366],{"className":3554},[365],[26,3556],{"className":3557,"style":358},[54],[26,3559],{"className":3560,"style":358},[54],[26,3562,3564,3567,3570,3610,3613,3617,3620,3660,3663,3666,3669,3672],{"className":3563},[38],[26,3565],{"className":3566,"style":1461},[42],[26,3568,3488],{"className":3569},[47,179],[26,3571,3573,3576],{"className":3572},[47],[26,3574,180],{"className":3575},[47,179],[26,3577,3579],{"className":3578},[184],[26,3580,3582,3602],{"className":3581},[188,189],[26,3583,3585,3599],{"className":3584},[193],[26,3586,3588],{"className":3587,"style":198},[197],[26,3589,3590,3593],{"style":201},[26,3591],{"className":3592,"style":206},[205],[26,3594,3596],{"className":3595},[210,211,212,213],[26,3597,158],{"className":3598},[47,48,213],[26,3600,221],{"className":3601},[220],[26,3603,3605],{"className":3604},[193],[26,3606,3608],{"className":3607,"style":228},[197],[26,3609],{},[26,3611,524],{"className":3612},[523],[26,3614],{"className":3615,"style":3616},[54],"margin-right:2em;",[26,3618],{"className":3619,"style":392},[54],[26,3621,3623,3626],{"className":3622},[47],[26,3624,251],{"className":3625},[47,179],[26,3627,3629],{"className":3628},[184],[26,3630,3632,3652],{"className":3631},[188,189],[26,3633,3635,3649],{"className":3634},[193],[26,3636,3638],{"className":3637,"style":198},[197],[26,3639,3640,3643],{"style":201},[26,3641],{"className":3642,"style":206},[205],[26,3644,3646],{"className":3645},[210,211,212,213],[26,3647,158],{"className":3648},[47,48,213],[26,3650,221],{"className":3651},[220],[26,3653,3655],{"className":3654},[193],[26,3656,3658],{"className":3657,"style":228},[197],[26,3659],{},[26,3661],{"className":3662,"style":358},[54],[26,3664],{"className":3665,"style":358},[54],[26,3667,366],{"className":3668},[365],[26,3670],{"className":3671,"style":358},[54],[26,3673],{"className":3674,"style":358},[54],[26,3676,3678,3681,3684,3687,3690,3803,3806,3809,3812,3815,3855,3858,3861,3864,3867],{"className":3677},[38],[26,3679],{"className":3680,"style":379},[42],[26,3682,384],{"className":3683,"style":383},[47,48],[26,3685],{"className":3686,"style":388},[54],[26,3688],{"className":3689,"style":392},[54],[26,3691,3693,3696,3699,3748,3751,3754,3757,3760,3800],{"className":3692},[396],[26,3694,403],{"className":3695,"style":402},[400,401],[26,3697,407],{"className":3698},[47,179],[26,3700,3702,3705],{"className":3701},[47],[26,3703,251],{"className":3704},[47,179],[26,3706,3708],{"className":3707},[184],[26,3709,3711,3740],{"className":3710},[188,189],[26,3712,3714,3737],{"className":3713},[193],[26,3715,3717],{"className":3716,"style":264},[197],[26,3718,3719,3722],{"style":201},[26,3720],{"className":3721,"style":206},[205],[26,3723,3725],{"className":3724},[210,211,212,213],[26,3726,3728,3731,3734],{"className":3727},[47,213],[26,3729,158],{"className":3730},[47,48,213],[26,3732,60],{"className":3733},[59,213],[26,3735,74],{"className":3736},[47,213],[26,3738,221],{"className":3739},[220],[26,3741,3743],{"className":3742},[193],[26,3744,3746],{"className":3745,"style":294},[197],[26,3747],{},[26,3749],{"className":3750,"style":55},[54],[26,3752,463],{"className":3753},[59],[26,3755],{"className":3756,"style":55},[54],[26,3758,471],{"className":3759,"style":470},[47,179],[26,3761,3763,3766],{"className":3762},[47],[26,3764,3511],{"className":3765},[47,179],[26,3767,3769],{"className":3768},[184],[26,3770,3772,3792],{"className":3771},[188,189],[26,3773,3775,3789],{"className":3774},[193],[26,3776,3778],{"className":3777,"style":198},[197],[26,3779,3780,3783],{"style":201},[26,3781],{"className":3782,"style":206},[205],[26,3784,3786],{"className":3785},[210,211,212,213],[26,3787,158],{"className":3788},[47,48,213],[26,3790,221],{"className":3791},[220],[26,3793,3795],{"className":3794},[193],[26,3796,3798],{"className":3797,"style":228},[197],[26,3799],{},[26,3801,516],{"className":3802,"style":402},[515,401],[26,3804],{"className":3805,"style":392},[54],[26,3807,524],{"className":3808},[523],[26,3810],{"className":3811,"style":3616},[54],[26,3813],{"className":3814,"style":392},[54],[26,3816,3818,3821],{"className":3817},[47],[26,3819,635],{"className":3820,"style":470},[47,179],[26,3822,3824],{"className":3823},[184],[26,3825,3827,3847],{"className":3826},[188,189],[26,3828,3830,3844],{"className":3829},[193],[26,3831,3833],{"className":3832,"style":198},[197],[26,3834,3835,3838],{"style":650},[26,3836],{"className":3837,"style":206},[205],[26,3839,3841],{"className":3840},[210,211,212,213],[26,3842,158],{"className":3843},[47,48,213],[26,3845,221],{"className":3846},[220],[26,3848,3850],{"className":3849},[193],[26,3851,3853],{"className":3852,"style":228},[197],[26,3854],{},[26,3856],{"className":3857,"style":358},[54],[26,3859],{"className":3860,"style":358},[54],[26,3862,366],{"className":3863},[365],[26,3865],{"className":3866,"style":358},[54],[26,3868],{"className":3869,"style":358},[54],[26,3871,3873,3876,3882,3885,3888,3940,3943],{"className":3872},[38],[26,3874],{"className":3875,"style":379},[42],[26,3877,3879],{"className":3878},[47],[26,3880,699],{"className":3881},[47,698],[26,3883],{"className":3884,"style":388},[54],[26,3886],{"className":3887,"style":392},[54],[26,3889,3891,3894,3897,3937],{"className":3890},[396],[26,3892,403],{"className":3893,"style":402},[400,401],[26,3895,715],{"className":3896,"style":470},[47,179],[26,3898,3900,3903],{"className":3899},[47],[26,3901,251],{"className":3902},[47,179],[26,3904,3906],{"className":3905},[184],[26,3907,3909,3929],{"className":3908},[188,189],[26,3910,3912,3926],{"className":3911},[193],[26,3913,3915],{"className":3914,"style":198},[197],[26,3916,3917,3920],{"style":201},[26,3918],{"className":3919,"style":206},[205],[26,3921,3923],{"className":3922},[210,211,212,213],[26,3924,158],{"className":3925},[47,48,213],[26,3927,221],{"className":3928},[220],[26,3930,3932],{"className":3931},[193],[26,3933,3935],{"className":3934,"style":228},[197],[26,3936],{},[26,3938,516],{"className":3939,"style":402},[515,401],[26,3941],{"className":3942,"style":392},[54],[26,3944,765],{"className":3945},[47],[11,3947,3948,3949,4001,4002,4019,4020,4035,4036,4309],{},"The output ",[26,3950,3952],{"className":3951},[29],[26,3953,3955],{"className":3954,"ariaHidden":34},[33],[26,3956,3958,3961],{"className":3957},[38],[26,3959],{"className":3960,"style":628},[42],[26,3962,3964,3967],{"className":3963},[47],[26,3965,635],{"className":3966,"style":470},[47,179],[26,3968,3970],{"className":3969},[184],[26,3971,3973,3993],{"className":3972},[188,189],[26,3974,3976,3990],{"className":3975},[193],[26,3977,3979],{"className":3978,"style":198},[197],[26,3980,3981,3984],{"style":650},[26,3982],{"className":3983,"style":206},[205],[26,3985,3987],{"className":3986},[210,211,212,213],[26,3988,158],{"className":3989},[47,48,213],[26,3991,221],{"className":3992},[220],[26,3994,3996],{"className":3995},[193],[26,3997,3999],{"className":3998,"style":228},[197],[26,4000],{}," is a probability vector over the vocabulary; its ",[26,4003,4005],{"className":4004},[29],[26,4006,4008],{"className":4007,"ariaHidden":34},[33],[26,4009,4011,4015],{"className":4010},[38],[26,4012],{"className":4013,"style":4014},[42],"height:0.6595em;",[26,4016,4018],{"className":4017},[47,48],"i","-th component is\nthe model's estimate that word ",[26,4021,4023],{"className":4022},[29],[26,4024,4026],{"className":4025,"ariaHidden":34},[33],[26,4027,4029,4032],{"className":4028},[38],[26,4030],{"className":4031,"style":4014},[42],[26,4033,4018],{"className":4034},[47,48]," comes next, ",[26,4037,4039],{"className":4038},[29],[26,4040,4042,4117,4136,4252],{"className":4041,"ariaHidden":34},[33],[26,4043,4045,4048,4053,4056,4108,4111,4114],{"className":4044},[38],[26,4046],{"className":4047,"style":379},[42],[26,4049,4052],{"className":4050,"style":4051},[47,48],"margin-right:0.1389em;","P",[26,4054,403],{"className":4055},[400],[26,4057,4059,4064],{"className":4058},[47],[26,4060,4063],{"className":4061,"style":4062},[47,48],"margin-right:0.0269em;","w",[26,4065,4067],{"className":4066},[184],[26,4068,4070,4100],{"className":4069},[188,189],[26,4071,4073,4097],{"className":4072},[193],[26,4074,4076],{"className":4075,"style":264},[197],[26,4077,4079,4082],{"style":4078},"top:-2.55em;margin-left:-0.0269em;margin-right:0.05em;",[26,4080],{"className":4081,"style":206},[205],[26,4083,4085],{"className":4084},[210,211,212,213],[26,4086,4088,4091,4094],{"className":4087},[47,213],[26,4089,158],{"className":4090},[47,48,213],[26,4092,463],{"className":4093},[59,213],[26,4095,74],{"className":4096},[47,213],[26,4098,221],{"className":4099},[220],[26,4101,4103],{"className":4102},[193],[26,4104,4106],{"className":4105,"style":294},[197],[26,4107],{},[26,4109],{"className":4110,"style":358},[54],[26,4112,366],{"className":4113},[365],[26,4115],{"className":4116,"style":358},[54],[26,4118,4120,4123,4126,4129,4133],{"className":4119},[38],[26,4121],{"className":4122,"style":379},[42],[26,4124,4018],{"className":4125},[47,48],[26,4127],{"className":4128,"style":358},[54],[26,4130,4132],{"className":4131},[365],"∣",[26,4134],{"className":4135,"style":358},[54],[26,4137,4139,4142,4182,4185,4188,4191,4194,4197,4200,4240,4243,4246,4249],{"className":4138},[38],[26,4140],{"className":4141,"style":379},[42],[26,4143,4145,4148],{"className":4144},[47],[26,4146,4063],{"className":4147,"style":4062},[47,48],[26,4149,4151],{"className":4150},[184],[26,4152,4154,4174],{"className":4153},[188,189],[26,4155,4157,4171],{"className":4156},[193],[26,4158,4160],{"className":4159,"style":264},[197],[26,4161,4162,4165],{"style":4078},[26,4163],{"className":4164,"style":206},[205],[26,4166,4168],{"className":4167},[210,211,212,213],[26,4169,74],{"className":4170},[47,213],[26,4172,221],{"className":4173},[220],[26,4175,4177],{"className":4176},[193],[26,4178,4180],{"className":4179,"style":228},[197],[26,4181],{},[26,4183,524],{"className":4184},[523],[26,4186],{"className":4187,"style":392},[54],[26,4189,1511],{"className":4190},[396],[26,4192],{"className":4193,"style":392},[54],[26,4195,524],{"className":4196},[523],[26,4198],{"className":4199,"style":392},[54],[26,4201,4203,4206],{"className":4202},[47],[26,4204,4063],{"className":4205,"style":4062},[47,48],[26,4207,4209],{"className":4208},[184],[26,4210,4212,4232],{"className":4211},[188,189],[26,4213,4215,4229],{"className":4214},[193],[26,4216,4218],{"className":4217,"style":198},[197],[26,4219,4220,4223],{"style":4078},[26,4221],{"className":4222,"style":206},[205],[26,4224,4226],{"className":4225},[210,211,212,213],[26,4227,158],{"className":4228},[47,48,213],[26,4230,221],{"className":4231},[220],[26,4233,4235],{"className":4234},[193],[26,4236,4238],{"className":4237,"style":228},[197],[26,4239],{},[26,4241,516],{"className":4242},[515],[26,4244],{"className":4245,"style":358},[54],[26,4247,366],{"className":4248},[365],[26,4250],{"className":4251,"style":358},[54],[26,4253,4255,4258,4298,4302,4305],{"className":4254},[38],[26,4256],{"className":4257,"style":379},[42],[26,4259,4261,4264],{"className":4260},[47],[26,4262,635],{"className":4263,"style":470},[47,179],[26,4265,4267],{"className":4266},[184],[26,4268,4270,4290],{"className":4269},[188,189],[26,4271,4273,4287],{"className":4272},[193],[26,4274,4276],{"className":4275,"style":198},[197],[26,4277,4278,4281],{"style":650},[26,4279],{"className":4280,"style":206},[205],[26,4282,4284],{"className":4283},[210,211,212,213],[26,4285,158],{"className":4286},[47,48,213],[26,4288,221],{"className":4289},[220],[26,4291,4293],{"className":4292},[193],[26,4294,4296],{"className":4295,"style":228},[197],[26,4297],{},[26,4299,4301],{"className":4300},[400],"[",[26,4303,4018],{"className":4304},[47,48],[26,4306,4308],{"className":4307},[515],"]",". The probability of a whole sequence is the product across positions,",[26,4311,4313],{"className":4312},[301],[26,4314,4316],{"className":4315},[29],[26,4317,4319,4399,4542,4621],{"className":4318,"ariaHidden":34},[33],[26,4320,4322,4325,4328,4331,4381,4384,4387,4390,4393,4396],{"className":4321},[38],[26,4323],{"className":4324,"style":379},[42],[26,4326,4052],{"className":4327,"style":4051},[47,48],[26,4329,403],{"className":4330},[400],[26,4332,4334,4337],{"className":4333},[47],[26,4335,4063],{"className":4336,"style":4062},[47,48],[26,4338,4340],{"className":4339},[184],[26,4341,4343,4373],{"className":4342},[188,189],[26,4344,4346,4370],{"className":4345},[193],[26,4347,4349],{"className":4348,"style":264},[197],[26,4350,4351,4354],{"style":4078},[26,4352],{"className":4353,"style":206},[205],[26,4355,4357],{"className":4356},[210,211,212,213],[26,4358,4360,4363,4367],{"className":4359},[47,213],[26,4361,74],{"className":4362},[47,213],[26,4364,4366],{"className":4365},[365,213],":",[26,4368,1551],{"className":4369},[47,48,213],[26,4371,221],{"className":4372},[220],[26,4374,4376],{"className":4375},[193],[26,4377,4379],{"className":4378,"style":228},[197],[26,4380],{},[26,4382,516],{"className":4383},[515],[26,4385],{"className":4386,"style":358},[54],[26,4388],{"className":4389,"style":358},[54],[26,4391,366],{"className":4392},[365],[26,4394],{"className":4395,"style":358},[54],[26,4397],{"className":4398,"style":358},[54],[26,4400,4402,4406,4483,4486,4489,4492,4533,4536,4539],{"className":4401},[38],[26,4403],{"className":4404,"style":4405},[42],"height:2.9291em;vertical-align:-1.2777em;",[26,4407,4410],{"className":4408},[561,4409],"op-limits",[26,4411,4413,4474],{"className":4412},[188,189],[26,4414,4416,4471],{"className":4415},[193],[26,4417,4420,4442,4456],{"className":4418,"style":4419},[197],"height:1.6514em;",[26,4421,4423,4427],{"style":4422},"top:-1.8723em;margin-left:0em;",[26,4424],{"className":4425,"style":4426},[205],"height:3.05em;",[26,4428,4430],{"className":4429},[210,211,212,213],[26,4431,4433,4436,4439],{"className":4432},[47,213],[26,4434,4018],{"className":4435},[47,48,213],[26,4437,366],{"className":4438},[365,213],[26,4440,74],{"className":4441},[47,213],[26,4443,4445,4448],{"style":4444},"top:-3.05em;",[26,4446],{"className":4447,"style":4426},[205],[26,4449,4450],{},[26,4451,4455],{"className":4452},[561,4453,4454],"op-symbol","large-op","∏",[26,4457,4459,4462],{"style":4458},"top:-4.3em;margin-left:0em;",[26,4460],{"className":4461,"style":4426},[205],[26,4463,4465],{"className":4464},[210,211,212,213],[26,4466,4468],{"className":4467},[47,213],[26,4469,1551],{"className":4470},[47,48,213],[26,4472,221],{"className":4473},[220],[26,4475,4477],{"className":4476},[193],[26,4478,4481],{"className":4479,"style":4480},[197],"height:1.2777em;",[26,4482],{},[26,4484],{"className":4485,"style":392},[54],[26,4487,4052],{"className":4488,"style":4051},[47,48],[26,4490,403],{"className":4491},[400],[26,4493,4495,4498],{"className":4494},[47],[26,4496,4063],{"className":4497,"style":4062},[47,48],[26,4499,4501],{"className":4500},[184],[26,4502,4504,4525],{"className":4503},[188,189],[26,4505,4507,4522],{"className":4506},[193],[26,4508,4511],{"className":4509,"style":4510},[197],"height:0.3117em;",[26,4512,4513,4516],{"style":4078},[26,4514],{"className":4515,"style":206},[205],[26,4517,4519],{"className":4518},[210,211,212,213],[26,4520,4018],{"className":4521},[47,48,213],[26,4523,221],{"className":4524},[220],[26,4526,4528],{"className":4527},[193],[26,4529,4531],{"className":4530,"style":228},[197],[26,4532],{},[26,4534],{"className":4535,"style":358},[54],[26,4537,4132],{"className":4538},[365],[26,4540],{"className":4541,"style":358},[54],[26,4543,4545,4548,4603,4606,4609,4612,4615,4618],{"className":4544},[38],[26,4546],{"className":4547,"style":379},[42],[26,4549,4551,4554],{"className":4550},[47],[26,4552,4063],{"className":4553,"style":4062},[47,48],[26,4555,4557],{"className":4556},[184],[26,4558,4560,4595],{"className":4559},[188,189],[26,4561,4563,4592],{"className":4562},[193],[26,4564,4566],{"className":4565,"style":4510},[197],[26,4567,4568,4571],{"style":4078},[26,4569],{"className":4570,"style":206},[205],[26,4572,4574],{"className":4573},[210,211,212,213],[26,4575,4577,4580,4583,4586,4589],{"className":4576},[47,213],[26,4578,74],{"className":4579},[47,213],[26,4581,4366],{"className":4582},[365,213],[26,4584,4018],{"className":4585},[47,48,213],[26,4587,60],{"className":4588},[59,213],[26,4590,74],{"className":4591},[47,213],[26,4593,221],{"className":4594},[220],[26,4596,4598],{"className":4597},[193],[26,4599,4601],{"className":4600,"style":294},[197],[26,4602],{},[26,4604,516],{"className":4605},[515],[26,4607],{"className":4608,"style":358},[54],[26,4610],{"className":4611,"style":358},[54],[26,4613,366],{"className":4614},[365],[26,4616],{"className":4617,"style":358},[54],[26,4619],{"className":4620,"style":358},[54],[26,4622,4624,4627,4694,4697,4737,4740,4780,4783],{"className":4623},[38],[26,4625],{"className":4626,"style":4405},[42],[26,4628,4630],{"className":4629},[561,4409],[26,4631,4633,4686],{"className":4632},[188,189],[26,4634,4636,4683],{"className":4635},[193],[26,4637,4639,4659,4669],{"className":4638,"style":4419},[197],[26,4640,4641,4644],{"style":4422},[26,4642],{"className":4643,"style":4426},[205],[26,4645,4647],{"className":4646},[210,211,212,213],[26,4648,4650,4653,4656],{"className":4649},[47,213],[26,4651,4018],{"className":4652},[47,48,213],[26,4654,366],{"className":4655},[365,213],[26,4657,74],{"className":4658},[47,213],[26,4660,4661,4664],{"style":4444},[26,4662],{"className":4663,"style":4426},[205],[26,4665,4666],{},[26,4667,4455],{"className":4668},[561,4453,4454],[26,4670,4671,4674],{"style":4458},[26,4672],{"className":4673,"style":4426},[205],[26,4675,4677],{"className":4676},[210,211,212,213],[26,4678,4680],{"className":4679},[47,213],[26,4681,1551],{"className":4682},[47,48,213],[26,4684,221],{"className":4685},[220],[26,4687,4689],{"className":4688},[193],[26,4690,4692],{"className":4691,"style":4480},[197],[26,4693],{},[26,4695],{"className":4696,"style":392},[54],[26,4698,4700,4703],{"className":4699},[47],[26,4701,635],{"className":4702,"style":470},[47,179],[26,4704,4706],{"className":4705},[184],[26,4707,4709,4729],{"className":4708},[188,189],[26,4710,4712,4726],{"className":4711},[193],[26,4713,4715],{"className":4714,"style":4510},[197],[26,4716,4717,4720],{"style":650},[26,4718],{"className":4719,"style":206},[205],[26,4721,4723],{"className":4722},[210,211,212,213],[26,4724,4018],{"className":4725},[47,48,213],[26,4727,221],{"className":4728},[220],[26,4730,4732],{"className":4731},[193],[26,4733,4735],{"className":4734,"style":228},[197],[26,4736],{},[26,4738,4301],{"className":4739},[400],[26,4741,4743,4746],{"className":4742},[47],[26,4744,4063],{"className":4745,"style":4062},[47,48],[26,4747,4749],{"className":4748},[184],[26,4750,4752,4772],{"className":4751},[188,189],[26,4753,4755,4769],{"className":4754},[193],[26,4756,4758],{"className":4757,"style":4510},[197],[26,4759,4760,4763],{"style":4078},[26,4761],{"className":4762,"style":206},[205],[26,4764,4766],{"className":4765},[210,211,212,213],[26,4767,4018],{"className":4768},[47,48,213],[26,4770,221],{"className":4771},[220],[26,4773,4775],{"className":4774},[193],[26,4776,4778],{"className":4777,"style":228},[197],[26,4779],{},[26,4781,4308],{"className":4782},[515],[26,4784,524],{"className":4785},[523],[11,4787,527,4788,4886,4887,4889,4890,4942,4943,765],{},[26,4789,4791],{"className":4790},[29],[26,4792,4794],{"className":4793,"ariaHidden":34},[33],[26,4795,4797,4800,4840,4843,4883],{"className":4796},[38],[26,4798],{"className":4799,"style":379},[42],[26,4801,4803,4806],{"className":4802},[47],[26,4804,635],{"className":4805,"style":470},[47,179],[26,4807,4809],{"className":4808},[184],[26,4810,4812,4832],{"className":4811},[188,189],[26,4813,4815,4829],{"className":4814},[193],[26,4816,4818],{"className":4817,"style":4510},[197],[26,4819,4820,4823],{"style":650},[26,4821],{"className":4822,"style":206},[205],[26,4824,4826],{"className":4825},[210,211,212,213],[26,4827,4018],{"className":4828},[47,48,213],[26,4830,221],{"className":4831},[220],[26,4833,4835],{"className":4834},[193],[26,4836,4838],{"className":4837,"style":228},[197],[26,4839],{},[26,4841,4301],{"className":4842},[400],[26,4844,4846,4849],{"className":4845},[47],[26,4847,4063],{"className":4848,"style":4062},[47,48],[26,4850,4852],{"className":4851},[184],[26,4853,4855,4875],{"className":4854},[188,189],[26,4856,4858,4872],{"className":4857},[193],[26,4859,4861],{"className":4860,"style":4510},[197],[26,4862,4863,4866],{"style":4078},[26,4864],{"className":4865,"style":206},[205],[26,4867,4869],{"className":4868},[210,211,212,213],[26,4870,4018],{"className":4871},[47,48,213],[26,4873,221],{"className":4874},[220],[26,4876,4878],{"className":4877},[193],[26,4879,4881],{"className":4880,"style":228},[197],[26,4882],{},[26,4884,4308],{"className":4885},[515]," denotes the probability the model placed on the ",[94,4888,34],{}," word ",[26,4891,4893],{"className":4892},[29],[26,4894,4896],{"className":4895,"ariaHidden":34},[33],[26,4897,4899,4902],{"className":4898},[38],[26,4900],{"className":4901,"style":2334},[42],[26,4903,4905,4908],{"className":4904},[47],[26,4906,4063],{"className":4907,"style":4062},[47,48],[26,4909,4911],{"className":4910},[184],[26,4912,4914,4934],{"className":4913},[188,189],[26,4915,4917,4931],{"className":4916},[193],[26,4918,4920],{"className":4919,"style":4510},[197],[26,4921,4922,4925],{"style":4078},[26,4923],{"className":4924,"style":206},[205],[26,4926,4928],{"className":4927},[210,211,212,213],[26,4929,4018],{"className":4930},[47,48,213],[26,4932,221],{"className":4933},[220],[26,4935,4937],{"className":4936},[193],[26,4938,4940],{"className":4939,"style":228},[197],[26,4941],{}," at\nstep ",[26,4944,4946],{"className":4945},[29],[26,4947,4949],{"className":4948,"ariaHidden":34},[33],[26,4950,4952,4955],{"className":4951},[38],[26,4953],{"className":4954,"style":4014},[42],[26,4956,4018],{"className":4957},[47,48],[1963,4959,4961],{"id":4960},"training-and-teacher-forcing","Training and teacher forcing",[11,4963,4964,4965,5027,5028,5043],{},"The RNN is trained to minimize the cross-entropy of the true next word. Because the target\ndistribution is a one-hot vector on the correct word ",[26,4966,4968],{"className":4967},[29],[26,4969,4971],{"className":4970,"ariaHidden":34},[33],[26,4972,4974,4978],{"className":4973},[38],[26,4975],{"className":4976,"style":4977},[42],"height:0.6389em;vertical-align:-0.2083em;",[26,4979,4981,4984],{"className":4980},[47],[26,4982,4063],{"className":4983,"style":4062},[47,48],[26,4985,4987],{"className":4986},[184],[26,4988,4990,5019],{"className":4989},[188,189],[26,4991,4993,5016],{"className":4992},[193],[26,4994,4996],{"className":4995,"style":264},[197],[26,4997,4998,5001],{"style":4078},[26,4999],{"className":5000,"style":206},[205],[26,5002,5004],{"className":5003},[210,211,212,213],[26,5005,5007,5010,5013],{"className":5006},[47,213],[26,5008,158],{"className":5009},[47,48,213],[26,5011,463],{"className":5012},[59,213],[26,5014,74],{"className":5015},[47,213],[26,5017,221],{"className":5018},[220],[26,5020,5022],{"className":5021},[193],[26,5023,5025],{"className":5024,"style":294},[197],[26,5026],{},", the cross-entropy at step ",[26,5029,5031],{"className":5030},[29],[26,5032,5034],{"className":5033,"ariaHidden":34},[33],[26,5035,5037,5040],{"className":5036},[38],[26,5038],{"className":5039,"style":154},[42],[26,5041,158],{"className":5042},[47,48],"\ncollapses to a single term, the negative log probability the model assigned to that word:",[26,5045,5047],{"className":5046},[301],[26,5048,5050],{"className":5049},[29],[26,5051,5053,5263],{"className":5052,"ariaHidden":34},[33],[26,5054,5056,5059,5108,5111,5199,5202,5205,5245,5248,5251,5254,5257,5260],{"className":5055},[38],[26,5057],{"className":5058,"style":379},[42],[26,5060,5062,5066],{"className":5061},[47],[26,5063,5065],{"className":5064},[47,48],"L",[26,5067,5069],{"className":5068},[184],[26,5070,5072,5100],{"className":5071},[188,189],[26,5073,5075,5097],{"className":5074},[193],[26,5076,5079],{"className":5077,"style":5078},[197],"height:0.3283em;",[26,5080,5081,5084],{"style":201},[26,5082],{"className":5083,"style":206},[205],[26,5085,5087],{"className":5086},[210,211,212,213],[26,5088,5090],{"className":5089},[47,213],[26,5091,5093],{"className":5092},[47,213],[26,5094,5096],{"className":5095},[47,698,213],"CE",[26,5098,221],{"className":5099},[220],[26,5101,5103],{"className":5102},[193],[26,5104,5106],{"className":5105,"style":228},[197],[26,5107],{},[26,5109,403],{"className":5110},[400],[26,5112,5114,5165],{"className":5113},[47],[26,5115,5118],{"className":5116},[47,5117],"accent",[26,5119,5121,5156],{"className":5120},[188,189],[26,5122,5124,5153],{"className":5123},[193],[26,5125,5128,5138],{"className":5126,"style":5127},[197],"height:0.7079em;",[26,5129,5131,5135],{"style":5130},"top:-3em;",[26,5132],{"className":5133,"style":5134},[205],"height:3em;",[26,5136,635],{"className":5137,"style":470},[47,179],[26,5139,5141,5144],{"style":5140},"top:-3.0134em;",[26,5142],{"className":5143,"style":5134},[205],[26,5145,5149],{"className":5146,"style":5148},[5147],"accent-body","left:-0.25em;",[26,5150,5152],{"className":5151},[47],"^",[26,5154,221],{"className":5155},[220],[26,5157,5159],{"className":5158},[193],[26,5160,5163],{"className":5161,"style":5162},[197],"height:0.1944em;",[26,5164],{},[26,5166,5168],{"className":5167},[184],[26,5169,5171,5191],{"className":5170},[188,189],[26,5172,5174,5188],{"className":5173},[193],[26,5175,5177],{"className":5176,"style":198},[197],[26,5178,5179,5182],{"style":650},[26,5180],{"className":5181,"style":206},[205],[26,5183,5185],{"className":5184},[210,211,212,213],[26,5186,158],{"className":5187},[47,48,213],[26,5189,221],{"className":5190},[220],[26,5192,5194],{"className":5193},[193],[26,5195,5197],{"className":5196,"style":228},[197],[26,5198],{},[26,5200,524],{"className":5201},[523],[26,5203],{"className":5204,"style":392},[54],[26,5206,5208,5211],{"className":5207},[47],[26,5209,635],{"className":5210,"style":470},[47,179],[26,5212,5214],{"className":5213},[184],[26,5215,5217,5237],{"className":5216},[188,189],[26,5218,5220,5234],{"className":5219},[193],[26,5221,5223],{"className":5222,"style":198},[197],[26,5224,5225,5228],{"style":650},[26,5226],{"className":5227,"style":206},[205],[26,5229,5231],{"className":5230},[210,211,212,213],[26,5232,158],{"className":5233},[47,48,213],[26,5235,221],{"className":5236},[220],[26,5238,5240],{"className":5239},[193],[26,5241,5243],{"className":5242,"style":228},[197],[26,5244],{},[26,5246,516],{"className":5247},[515],[26,5249],{"className":5250,"style":358},[54],[26,5252],{"className":5253,"style":358},[54],[26,5255,366],{"className":5256},[365],[26,5258],{"className":5259,"style":358},[54],[26,5261],{"className":5262,"style":358},[54],[26,5264,5266,5269,5272,5275,5283,5286,5365,5368,5417,5420],{"className":5265},[38],[26,5267],{"className":5268,"style":379},[42],[26,5270,60],{"className":5271},[47],[26,5273],{"className":5274,"style":392},[54],[26,5276,5278],{"className":5277},[561],[26,5279,5282],{"className":5280,"style":5281},[47,698],"margin-right:0.0139em;","log",[26,5284],{"className":5285,"style":392},[54],[26,5287,5289,5331],{"className":5288},[47],[26,5290,5292],{"className":5291},[47,5117],[26,5293,5295,5323],{"className":5294},[188,189],[26,5296,5298,5320],{"className":5297},[193],[26,5299,5301,5309],{"className":5300,"style":5127},[197],[26,5302,5303,5306],{"style":5130},[26,5304],{"className":5305,"style":5134},[205],[26,5307,635],{"className":5308,"style":470},[47,179],[26,5310,5311,5314],{"style":5140},[26,5312],{"className":5313,"style":5134},[205],[26,5315,5317],{"className":5316,"style":5148},[5147],[26,5318,5152],{"className":5319},[47],[26,5321,221],{"className":5322},[220],[26,5324,5326],{"className":5325},[193],[26,5327,5329],{"className":5328,"style":5162},[197],[26,5330],{},[26,5332,5334],{"className":5333},[184],[26,5335,5337,5357],{"className":5336},[188,189],[26,5338,5340,5354],{"className":5339},[193],[26,5341,5343],{"className":5342,"style":198},[197],[26,5344,5345,5348],{"style":650},[26,5346],{"className":5347,"style":206},[205],[26,5349,5351],{"className":5350},[210,211,212,213],[26,5352,158],{"className":5353},[47,48,213],[26,5355,221],{"className":5356},[220],[26,5358,5360],{"className":5359},[193],[26,5361,5363],{"className":5362,"style":228},[197],[26,5364],{},[26,5366,4301],{"className":5367},[400],[26,5369,5371,5374],{"className":5370},[47],[26,5372,4063],{"className":5373,"style":4062},[47,48],[26,5375,5377],{"className":5376},[184],[26,5378,5380,5409],{"className":5379},[188,189],[26,5381,5383,5406],{"className":5382},[193],[26,5384,5386],{"className":5385,"style":264},[197],[26,5387,5388,5391],{"style":4078},[26,5389],{"className":5390,"style":206},[205],[26,5392,5394],{"className":5393},[210,211,212,213],[26,5395,5397,5400,5403],{"className":5396},[47,213],[26,5398,158],{"className":5399},[47,48,213],[26,5401,463],{"className":5402},[59,213],[26,5404,74],{"className":5405},[47,213],[26,5407,221],{"className":5408},[220],[26,5410,5412],{"className":5411},[193],[26,5413,5415],{"className":5414,"style":294},[197],[26,5416],{},[26,5418,4308],{"className":5419},[515],[26,5421,765],{"className":5422},[47],[11,5424,5425,5426,5429,5430,5491,5492,5507,5508,5575,5576,5579],{},"The loss for a training sequence is the average of these over all positions. One subtlety\ngoverns training: at each step we feed the model the ",[21,5427,5428],{},"correct"," history ",[26,5431,5433],{"className":5432},[29],[26,5434,5436],{"className":5435,"ariaHidden":34},[33],[26,5437,5439,5442],{"className":5438},[38],[26,5440],{"className":5441,"style":2334},[42],[26,5443,5445,5448],{"className":5444},[47],[26,5446,4063],{"className":5447,"style":4062},[47,48],[26,5449,5451],{"className":5450},[184],[26,5452,5454,5483],{"className":5453},[188,189],[26,5455,5457,5480],{"className":5456},[193],[26,5458,5460],{"className":5459,"style":264},[197],[26,5461,5462,5465],{"style":4078},[26,5463],{"className":5464,"style":206},[205],[26,5466,5468],{"className":5467},[210,211,212,213],[26,5469,5471,5474,5477],{"className":5470},[47,213],[26,5472,74],{"className":5473},[47,213],[26,5475,4366],{"className":5476},[365,213],[26,5478,158],{"className":5479},[47,48,213],[26,5481,221],{"className":5482},[220],[26,5484,5486],{"className":5485},[193],[26,5487,5489],{"className":5488,"style":228},[197],[26,5490],{},", not the\nwords it actually predicted. If the model guesses wrong at step ",[26,5493,5495],{"className":5494},[29],[26,5496,5498],{"className":5497,"ariaHidden":34},[33],[26,5499,5501,5504],{"className":5500},[38],[26,5502],{"className":5503,"style":154},[42],[26,5505,158],{"className":5506},[47,48],", we ignore that guess and\nstill hand it the true ",[26,5509,5511],{"className":5510},[29],[26,5512,5514],{"className":5513,"ariaHidden":34},[33],[26,5515,5517,5520],{"className":5516},[38],[26,5518],{"className":5519,"style":4977},[42],[26,5521,5523,5526],{"className":5522},[47],[26,5524,4063],{"className":5525,"style":4062},[47,48],[26,5527,5529],{"className":5528},[184],[26,5530,5532,5567],{"className":5531},[188,189],[26,5533,5535,5564],{"className":5534},[193],[26,5536,5538],{"className":5537,"style":264},[197],[26,5539,5540,5543],{"style":4078},[26,5541],{"className":5542,"style":206},[205],[26,5544,5546],{"className":5545},[210,211,212,213],[26,5547,5549,5552,5555,5558,5561],{"className":5548},[47,213],[26,5550,74],{"className":5551},[47,213],[26,5553,4366],{"className":5554},[365,213],[26,5556,158],{"className":5557},[47,48,213],[26,5559,463],{"className":5560},[59,213],[26,5562,74],{"className":5563},[47,213],[26,5565,221],{"className":5566},[220],[26,5568,5570],{"className":5569},[193],[26,5571,5573],{"className":5572,"style":294},[197],[26,5574],{}," for the next step. Always supplying the gold prefix is called\n",[21,5577,5578],{},"teacher forcing",", and it keeps training stable by preventing early mistakes from cascading.",[1319,5581],{"hash":5582},"295cce872b423963a7529b2952ca9e7d33ae697b06f5c0aabeed6611d43da8b3",[11,5584,5585,5586,5589,5590,5642,5643,5646],{},"The same trained model can ",[21,5587,5588],{},"generate"," text: seed it with a start symbol, sample a word from\n",[26,5591,5593],{"className":5592},[29],[26,5594,5596],{"className":5595,"ariaHidden":34},[33],[26,5597,5599,5602],{"className":5598},[38],[26,5600],{"className":5601,"style":628},[42],[26,5603,5605,5608],{"className":5604},[47],[26,5606,635],{"className":5607,"style":470},[47,179],[26,5609,5611],{"className":5610},[184],[26,5612,5614,5634],{"className":5613},[188,189],[26,5615,5617,5631],{"className":5616},[193],[26,5618,5620],{"className":5619,"style":264},[197],[26,5621,5622,5625],{"style":650},[26,5623],{"className":5624,"style":206},[205],[26,5626,5628],{"className":5627},[210,211,212,213],[26,5629,74],{"className":5630},[47,213],[26,5632,221],{"className":5633},[220],[26,5635,5637],{"className":5636},[193],[26,5638,5640],{"className":5639,"style":228},[197],[26,5641],{},", feed that word back in as the next input, and repeat until an end symbol appears.\nBecause each generated word is fed back as the next step's input, this is called\n",[21,5644,5645],{},"autoregressive generation"," — the model conditions on its own choices, one word at a time.",[11,5648,5649,5650,5653,5654,5658,5659,5662,5663,5715,5716,765],{},"RNNs slot into other tasks with only the output layer changed. For ",[21,5651,5652],{},"sequence labeling"," — the\n",[15,5655,5657],{"href":5656},"\u002Fnatural-language-processing\u002Fsequences\u002Fsequence-labeling","part-of-speech and named-entity tagging","\nof the previous lesson — a softmax over the tagset at each step emits one label per token. For\n",[21,5660,5661],{},"sequence classification",", the final hidden state ",[26,5664,5666],{"className":5665},[29],[26,5667,5669],{"className":5668,"ariaHidden":34},[33],[26,5670,5672,5675],{"className":5671},[38],[26,5673],{"className":5674,"style":314},[42],[26,5676,5678,5681],{"className":5677},[47],[26,5679,251],{"className":5680},[47,179],[26,5682,5684],{"className":5683},[184],[26,5685,5687,5707],{"className":5686},[188,189],[26,5688,5690,5704],{"className":5689},[193],[26,5691,5693],{"className":5692,"style":1539},[197],[26,5694,5695,5698],{"style":201},[26,5696],{"className":5697,"style":206},[205],[26,5699,5701],{"className":5700},[210,211,212,213],[26,5702,1551],{"className":5703},[47,48,213],[26,5705,221],{"className":5706},[220],[26,5708,5710],{"className":5709},[193],[26,5711,5713],{"className":5712,"style":228},[197],[26,5714],{}," serves as a compressed\nsummary of the whole text and is fed to a feedforward classifier; the classification loss then\nbackpropagates all the way through the RNN, an instance of ",[21,5717,5718],{},"end-to-end training",[127,5720,5722],{"id":5721},"backpropagation-through-time","Backpropagation through time",[11,5724,5725,5726,5778,5779,5782,5783,5844,5845,5906,5907,5959,5960,5963,5964],{},"Training an unrolled RNN is ordinary backpropagation applied to the unrolled graph, with one\nnew complication: a hidden state ",[26,5727,5729],{"className":5728},[29],[26,5730,5732],{"className":5731,"ariaHidden":34},[33],[26,5733,5735,5738],{"className":5734},[38],[26,5736],{"className":5737,"style":314},[42],[26,5739,5741,5744],{"className":5740},[47],[26,5742,251],{"className":5743},[47,179],[26,5745,5747],{"className":5746},[184],[26,5748,5750,5770],{"className":5749},[188,189],[26,5751,5753,5767],{"className":5752},[193],[26,5754,5756],{"className":5755,"style":198},[197],[26,5757,5758,5761],{"style":201},[26,5759],{"className":5760,"style":206},[205],[26,5762,5764],{"className":5763},[210,211,212,213],[26,5765,158],{"className":5766},[47,48,213],[26,5768,221],{"className":5769},[220],[26,5771,5773],{"className":5772},[193],[26,5774,5776],{"className":5775,"style":228},[197],[26,5777],{}," influences the loss at ",[94,5780,5781],{},"every"," later step,\nbecause it feeds ",[26,5784,5786],{"className":5785},[29],[26,5787,5789],{"className":5788,"ariaHidden":34},[33],[26,5790,5792,5795],{"className":5791},[38],[26,5793],{"className":5794,"style":244},[42],[26,5796,5798,5801],{"className":5797},[47],[26,5799,251],{"className":5800},[47,179],[26,5802,5804],{"className":5803},[184],[26,5805,5807,5836],{"className":5806},[188,189],[26,5808,5810,5833],{"className":5809},[193],[26,5811,5813],{"className":5812,"style":264},[197],[26,5814,5815,5818],{"style":201},[26,5816],{"className":5817,"style":206},[205],[26,5819,5821],{"className":5820},[210,211,212,213],[26,5822,5824,5827,5830],{"className":5823},[47,213],[26,5825,158],{"className":5826},[47,48,213],[26,5828,463],{"className":5829},[59,213],[26,5831,74],{"className":5832},[47,213],[26,5834,221],{"className":5835},[220],[26,5837,5839],{"className":5838},[193],[26,5840,5842],{"className":5841,"style":294},[197],[26,5843],{},", which feeds ",[26,5846,5848],{"className":5847},[29],[26,5849,5851],{"className":5850,"ariaHidden":34},[33],[26,5852,5854,5857],{"className":5853},[38],[26,5855],{"className":5856,"style":244},[42],[26,5858,5860,5863],{"className":5859},[47],[26,5861,251],{"className":5862},[47,179],[26,5864,5866],{"className":5865},[184],[26,5867,5869,5898],{"className":5868},[188,189],[26,5870,5872,5895],{"className":5871},[193],[26,5873,5875],{"className":5874,"style":264},[197],[26,5876,5877,5880],{"style":201},[26,5878],{"className":5879,"style":206},[205],[26,5881,5883],{"className":5882},[210,211,212,213],[26,5884,5886,5889,5892],{"className":5885},[47,213],[26,5887,158],{"className":5888},[47,48,213],[26,5890,463],{"className":5891},[59,213],[26,5893,1282],{"className":5894},[47,213],[26,5896,221],{"className":5897},[220],[26,5899,5901],{"className":5900},[193],[26,5902,5904],{"className":5903,"style":294},[197],[26,5905],{},", and so on. To find the\ngradient reaching ",[26,5908,5910],{"className":5909},[29],[26,5911,5913],{"className":5912,"ariaHidden":34},[33],[26,5914,5916,5919],{"className":5915},[38],[26,5917],{"className":5918,"style":314},[42],[26,5920,5922,5925],{"className":5921},[47],[26,5923,251],{"className":5924},[47,179],[26,5926,5928],{"className":5927},[184],[26,5929,5931,5951],{"className":5930},[188,189],[26,5932,5934,5948],{"className":5933},[193],[26,5935,5937],{"className":5936,"style":198},[197],[26,5938,5939,5942],{"style":201},[26,5940],{"className":5941,"style":206},[205],[26,5943,5945],{"className":5944},[210,211,212,213],[26,5946,158],{"className":5947},[47,48,213],[26,5949,221],{"className":5950},[220],[26,5952,5954],{"className":5953},[193],[26,5955,5957],{"className":5956,"style":228},[197],[26,5958],{}," we must sum the error flowing back from all of those futures.\nThe resulting two-pass algorithm — forward to compute and save every hidden state and loss,\nthen backward through time to accumulate gradients — is ",[21,5961,5962],{},"backpropagation through time","\n(BPTT).",[117,5965,5966],{},[15,5967,5971],{"href":5968,"ariaDescribedBy":5969,"dataFootnoteRef":6,"id":5970},"#user-content-fn-jm-bptt",[123],"user-content-fnref-jm-bptt","3",[11,5973,5974,5975,1075,5990,1075,6005,6020],{},"Because the weights are shared, each of ",[26,5976,5978],{"className":5977},[29],[26,5979,5981],{"className":5980,"ariaHidden":34},[33],[26,5982,5984,5987],{"className":5983},[38],[26,5985],{"className":5986,"style":576},[42],[26,5988,407],{"className":5989},[47,179],[26,5991,5993],{"className":5992},[29],[26,5994,5996],{"className":5995,"ariaHidden":34},[33],[26,5997,5999,6002],{"className":5998},[38],[26,6000],{"className":6001,"style":576},[42],[26,6003,715],{"className":6004,"style":470},[47,179],[26,6006,6008],{"className":6007},[29],[26,6009,6011],{"className":6010,"ariaHidden":34},[33],[26,6012,6014,6017],{"className":6013},[38],[26,6015],{"className":6016,"style":576},[42],[26,6018,471],{"className":6019,"style":470},[47,179]," receives a\ngradient contribution from every time step; the per-step gradients are summed before the update.\nThe mechanics are standard, but the shape of the backward pass has a consequence that governs\nwhether the network can actually learn long-range structure.",[1963,6022,6024],{"id":6023},"worked-example-unrolling-the-backward-chain","Worked example: unrolling the backward chain",[11,6026,6027,6028,6043,6044,6097,6098,6150,6151,6203,6204,6256,6257,6309,6310,6362],{},"Unroll three steps and follow one gradient path. The loss at step ",[26,6029,6031],{"className":6030},[29],[26,6032,6034],{"className":6033,"ariaHidden":34},[33],[26,6035,6037,6040],{"className":6036},[38],[26,6038],{"className":6039,"style":70},[42],[26,6041,5971],{"className":6042},[47],", call it ",[26,6045,6047],{"className":6046},[29],[26,6048,6050],{"className":6049,"ariaHidden":34},[33],[26,6051,6053,6057],{"className":6052},[38],[26,6054],{"className":6055,"style":6056},[42],"height:0.8333em;vertical-align:-0.15em;",[26,6058,6060,6063],{"className":6059},[47],[26,6061,5065],{"className":6062},[47,48],[26,6064,6066],{"className":6065},[184],[26,6067,6069,6089],{"className":6068},[188,189],[26,6070,6072,6086],{"className":6071},[193],[26,6073,6075],{"className":6074,"style":264},[197],[26,6076,6077,6080],{"style":201},[26,6078],{"className":6079,"style":206},[205],[26,6081,6083],{"className":6082},[210,211,212,213],[26,6084,5971],{"className":6085},[47,213],[26,6087,221],{"className":6088},[220],[26,6090,6092],{"className":6091},[193],[26,6093,6095],{"className":6094,"style":228},[197],[26,6096],{},",\ndepends on ",[26,6099,6101],{"className":6100},[29],[26,6102,6104],{"className":6103,"ariaHidden":34},[33],[26,6105,6107,6110],{"className":6106},[38],[26,6108],{"className":6109,"style":314},[42],[26,6111,6113,6116],{"className":6112},[47],[26,6114,251],{"className":6115},[47,179],[26,6117,6119],{"className":6118},[184],[26,6120,6122,6142],{"className":6121},[188,189],[26,6123,6125,6139],{"className":6124},[193],[26,6126,6128],{"className":6127,"style":264},[197],[26,6129,6130,6133],{"style":201},[26,6131],{"className":6132,"style":206},[205],[26,6134,6136],{"className":6135},[210,211,212,213],[26,6137,5971],{"className":6138},[47,213],[26,6140,221],{"className":6141},[220],[26,6143,6145],{"className":6144},[193],[26,6146,6148],{"className":6147,"style":228},[197],[26,6149],{},", which depends on ",[26,6152,6154],{"className":6153},[29],[26,6155,6157],{"className":6156,"ariaHidden":34},[33],[26,6158,6160,6163],{"className":6159},[38],[26,6161],{"className":6162,"style":314},[42],[26,6164,6166,6169],{"className":6165},[47],[26,6167,251],{"className":6168},[47,179],[26,6170,6172],{"className":6171},[184],[26,6173,6175,6195],{"className":6174},[188,189],[26,6176,6178,6192],{"className":6177},[193],[26,6179,6181],{"className":6180,"style":264},[197],[26,6182,6183,6186],{"style":201},[26,6184],{"className":6185,"style":206},[205],[26,6187,6189],{"className":6188},[210,211,212,213],[26,6190,1282],{"className":6191},[47,213],[26,6193,221],{"className":6194},[220],[26,6196,6198],{"className":6197},[193],[26,6199,6201],{"className":6200,"style":228},[197],[26,6202],{},", which depends on\n",[26,6205,6207],{"className":6206},[29],[26,6208,6210],{"className":6209,"ariaHidden":34},[33],[26,6211,6213,6216],{"className":6212},[38],[26,6214],{"className":6215,"style":314},[42],[26,6217,6219,6222],{"className":6218},[47],[26,6220,251],{"className":6221},[47,179],[26,6223,6225],{"className":6224},[184],[26,6226,6228,6248],{"className":6227},[188,189],[26,6229,6231,6245],{"className":6230},[193],[26,6232,6234],{"className":6233,"style":264},[197],[26,6235,6236,6239],{"style":201},[26,6237],{"className":6238,"style":206},[205],[26,6240,6242],{"className":6241},[210,211,212,213],[26,6243,74],{"className":6244},[47,213],[26,6246,221],{"className":6247},[220],[26,6249,6251],{"className":6250},[193],[26,6252,6254],{"className":6253,"style":228},[197],[26,6255],{},". The chain rule for the gradient of ",[26,6258,6260],{"className":6259},[29],[26,6261,6263],{"className":6262,"ariaHidden":34},[33],[26,6264,6266,6269],{"className":6265},[38],[26,6267],{"className":6268,"style":6056},[42],[26,6270,6272,6275],{"className":6271},[47],[26,6273,5065],{"className":6274},[47,48],[26,6276,6278],{"className":6277},[184],[26,6279,6281,6301],{"className":6280},[188,189],[26,6282,6284,6298],{"className":6283},[193],[26,6285,6287],{"className":6286,"style":264},[197],[26,6288,6289,6292],{"style":201},[26,6290],{"className":6291,"style":206},[205],[26,6293,6295],{"className":6294},[210,211,212,213],[26,6296,5971],{"className":6297},[47,213],[26,6299,221],{"className":6300},[220],[26,6302,6304],{"className":6303},[193],[26,6305,6307],{"className":6306,"style":228},[197],[26,6308],{}," with respect to the earliest\nhidden state ",[26,6311,6313],{"className":6312},[29],[26,6314,6316],{"className":6315,"ariaHidden":34},[33],[26,6317,6319,6322],{"className":6318},[38],[26,6320],{"className":6321,"style":314},[42],[26,6323,6325,6328],{"className":6324},[47],[26,6326,251],{"className":6327},[47,179],[26,6329,6331],{"className":6330},[184],[26,6332,6334,6354],{"className":6333},[188,189],[26,6335,6337,6351],{"className":6336},[193],[26,6338,6340],{"className":6339,"style":264},[197],[26,6341,6342,6345],{"style":201},[26,6343],{"className":6344,"style":206},[205],[26,6346,6348],{"className":6347},[210,211,212,213],[26,6349,74],{"className":6350},[47,213],[26,6352,221],{"className":6353},[220],[26,6355,6357],{"className":6356},[193],[26,6358,6360],{"className":6359,"style":228},[197],[26,6361],{}," is a product of per-step Jacobians:",[26,6364,6366],{"className":6365},[301],[26,6367,6369],{"className":6368},[29],[26,6370,6372,6541,7150],{"className":6371,"ariaHidden":34},[33],[26,6373,6375,6379,6532,6535,6538],{"className":6374},[38],[26,6376],{"className":6377,"style":6378},[42],"height:2.2074em;vertical-align:-0.836em;",[26,6380,6382,6386,6529],{"className":6381},[47],[26,6383],{"className":6384},[400,6385],"nulldelimiter",[26,6387,6390],{"className":6388},[6389],"mfrac",[26,6391,6393,6520],{"className":6392},[188,189],[26,6394,6396,6517],{"className":6395},[193],[26,6397,6400,6454,6465],{"className":6398,"style":6399},[197],"height:1.3714em;",[26,6401,6403,6406],{"style":6402},"top:-2.314em;",[26,6404],{"className":6405,"style":5134},[205],[26,6407,6409,6414],{"className":6408},[47],[26,6410,6413],{"className":6411,"style":6412},[47],"margin-right:0.0556em;","∂",[26,6415,6417,6420],{"className":6416},[47],[26,6418,251],{"className":6419},[47,179],[26,6421,6423],{"className":6422},[184],[26,6424,6426,6446],{"className":6425},[188,189],[26,6427,6429,6443],{"className":6428},[193],[26,6430,6432],{"className":6431,"style":264},[197],[26,6433,6434,6437],{"style":201},[26,6435],{"className":6436,"style":206},[205],[26,6438,6440],{"className":6439},[210,211,212,213],[26,6441,74],{"className":6442},[47,213],[26,6444,221],{"className":6445},[220],[26,6447,6449],{"className":6448},[193],[26,6450,6452],{"className":6451,"style":228},[197],[26,6453],{},[26,6455,6457,6460],{"style":6456},"top:-3.23em;",[26,6458],{"className":6459,"style":5134},[205],[26,6461],{"className":6462,"style":6464},[6463],"frac-line","border-bottom-width:0.04em;",[26,6466,6468,6471],{"style":6467},"top:-3.677em;",[26,6469],{"className":6470,"style":5134},[205],[26,6472,6474,6477],{"className":6473},[47],[26,6475,6413],{"className":6476,"style":6412},[47],[26,6478,6480,6483],{"className":6479},[47],[26,6481,5065],{"className":6482},[47,48],[26,6484,6486],{"className":6485},[184],[26,6487,6489,6509],{"className":6488},[188,189],[26,6490,6492,6506],{"className":6491},[193],[26,6493,6495],{"className":6494,"style":264},[197],[26,6496,6497,6500],{"style":201},[26,6498],{"className":6499,"style":206},[205],[26,6501,6503],{"className":6502},[210,211,212,213],[26,6504,5971],{"className":6505},[47,213],[26,6507,221],{"className":6508},[220],[26,6510,6512],{"className":6511},[193],[26,6513,6515],{"className":6514,"style":228},[197],[26,6516],{},[26,6518,221],{"className":6519},[220],[26,6521,6523],{"className":6522},[193],[26,6524,6527],{"className":6525,"style":6526},[197],"height:0.836em;",[26,6528],{},[26,6530],{"className":6531},[515,6385],[26,6533],{"className":6534,"style":358},[54],[26,6536,366],{"className":6537},[365],[26,6539],{"className":6540,"style":358},[54],[26,6542,6544,6548,6690,6693,6835,6838,6980,6983,6986,6989,7141,7144,7147],{"className":6543},[38],[26,6545],{"className":6546,"style":6547},[42],"height:2.2658em;vertical-align:-0.8943em;",[26,6549,6551,6554,6687],{"className":6550},[47],[26,6552],{"className":6553},[400,6385],[26,6555,6557],{"className":6556},[6389],[26,6558,6560,6679],{"className":6559},[188,189],[26,6561,6563,6676],{"className":6562},[193],[26,6564,6566,6617,6625],{"className":6565,"style":6399},[197],[26,6567,6568,6571],{"style":6402},[26,6569],{"className":6570,"style":5134},[205],[26,6572,6574,6577],{"className":6573},[47],[26,6575,6413],{"className":6576,"style":6412},[47],[26,6578,6580,6583],{"className":6579},[47],[26,6581,251],{"className":6582},[47,179],[26,6584,6586],{"className":6585},[184],[26,6587,6589,6609],{"className":6588},[188,189],[26,6590,6592,6606],{"className":6591},[193],[26,6593,6595],{"className":6594,"style":264},[197],[26,6596,6597,6600],{"style":201},[26,6598],{"className":6599,"style":206},[205],[26,6601,6603],{"className":6602},[210,211,212,213],[26,6604,5971],{"className":6605},[47,213],[26,6607,221],{"className":6608},[220],[26,6610,6612],{"className":6611},[193],[26,6613,6615],{"className":6614,"style":228},[197],[26,6616],{},[26,6618,6619,6622],{"style":6456},[26,6620],{"className":6621,"style":5134},[205],[26,6623],{"className":6624,"style":6464},[6463],[26,6626,6627,6630],{"style":6467},[26,6628],{"className":6629,"style":5134},[205],[26,6631,6633,6636],{"className":6632},[47],[26,6634,6413],{"className":6635,"style":6412},[47],[26,6637,6639,6642],{"className":6638},[47],[26,6640,5065],{"className":6641},[47,48],[26,6643,6645],{"className":6644},[184],[26,6646,6648,6668],{"className":6647},[188,189],[26,6649,6651,6665],{"className":6650},[193],[26,6652,6654],{"className":6653,"style":264},[197],[26,6655,6656,6659],{"style":201},[26,6657],{"className":6658,"style":206},[205],[26,6660,6662],{"className":6661},[210,211,212,213],[26,6663,5971],{"className":6664},[47,213],[26,6666,221],{"className":6667},[220],[26,6669,6671],{"className":6670},[193],[26,6672,6674],{"className":6673,"style":228},[197],[26,6675],{},[26,6677,221],{"className":6678},[220],[26,6680,6682],{"className":6681},[193],[26,6683,6685],{"className":6684,"style":6526},[197],[26,6686],{},[26,6688],{"className":6689},[515,6385],[26,6691],{"className":6692,"style":358},[54],[26,6694,6696,6699,6832],{"className":6695},[47],[26,6697],{"className":6698},[400,6385],[26,6700,6702],{"className":6701},[6389],[26,6703,6705,6824],{"className":6704},[188,189],[26,6706,6708,6821],{"className":6707},[193],[26,6709,6711,6762,6770],{"className":6710,"style":6399},[197],[26,6712,6713,6716],{"style":6402},[26,6714],{"className":6715,"style":5134},[205],[26,6717,6719,6722],{"className":6718},[47],[26,6720,6413],{"className":6721,"style":6412},[47],[26,6723,6725,6728],{"className":6724},[47],[26,6726,251],{"className":6727},[47,179],[26,6729,6731],{"className":6730},[184],[26,6732,6734,6754],{"className":6733},[188,189],[26,6735,6737,6751],{"className":6736},[193],[26,6738,6740],{"className":6739,"style":264},[197],[26,6741,6742,6745],{"style":201},[26,6743],{"className":6744,"style":206},[205],[26,6746,6748],{"className":6747},[210,211,212,213],[26,6749,1282],{"className":6750},[47,213],[26,6752,221],{"className":6753},[220],[26,6755,6757],{"className":6756},[193],[26,6758,6760],{"className":6759,"style":228},[197],[26,6761],{},[26,6763,6764,6767],{"style":6456},[26,6765],{"className":6766,"style":5134},[205],[26,6768],{"className":6769,"style":6464},[6463],[26,6771,6772,6775],{"style":6467},[26,6773],{"className":6774,"style":5134},[205],[26,6776,6778,6781],{"className":6777},[47],[26,6779,6413],{"className":6780,"style":6412},[47],[26,6782,6784,6787],{"className":6783},[47],[26,6785,251],{"className":6786},[47,179],[26,6788,6790],{"className":6789},[184],[26,6791,6793,6813],{"className":6792},[188,189],[26,6794,6796,6810],{"className":6795},[193],[26,6797,6799],{"className":6798,"style":264},[197],[26,6800,6801,6804],{"style":201},[26,6802],{"className":6803,"style":206},[205],[26,6805,6807],{"className":6806},[210,211,212,213],[26,6808,5971],{"className":6809},[47,213],[26,6811,221],{"className":6812},[220],[26,6814,6816],{"className":6815},[193],[26,6817,6819],{"className":6818,"style":228},[197],[26,6820],{},[26,6822,221],{"className":6823},[220],[26,6825,6827],{"className":6826},[193],[26,6828,6830],{"className":6829,"style":6526},[197],[26,6831],{},[26,6833],{"className":6834},[515,6385],[26,6836],{"className":6837,"style":358},[54],[26,6839,6841,6844,6977],{"className":6840},[47],[26,6842],{"className":6843},[400,6385],[26,6845,6847],{"className":6846},[6389],[26,6848,6850,6969],{"className":6849},[188,189],[26,6851,6853,6966],{"className":6852},[193],[26,6854,6856,6907,6915],{"className":6855,"style":6399},[197],[26,6857,6858,6861],{"style":6402},[26,6859],{"className":6860,"style":5134},[205],[26,6862,6864,6867],{"className":6863},[47],[26,6865,6413],{"className":6866,"style":6412},[47],[26,6868,6870,6873],{"className":6869},[47],[26,6871,251],{"className":6872},[47,179],[26,6874,6876],{"className":6875},[184],[26,6877,6879,6899],{"className":6878},[188,189],[26,6880,6882,6896],{"className":6881},[193],[26,6883,6885],{"className":6884,"style":264},[197],[26,6886,6887,6890],{"style":201},[26,6888],{"className":6889,"style":206},[205],[26,6891,6893],{"className":6892},[210,211,212,213],[26,6894,74],{"className":6895},[47,213],[26,6897,221],{"className":6898},[220],[26,6900,6902],{"className":6901},[193],[26,6903,6905],{"className":6904,"style":228},[197],[26,6906],{},[26,6908,6909,6912],{"style":6456},[26,6910],{"className":6911,"style":5134},[205],[26,6913],{"className":6914,"style":6464},[6463],[26,6916,6917,6920],{"style":6467},[26,6918],{"className":6919,"style":5134},[205],[26,6921,6923,6926],{"className":6922},[47],[26,6924,6413],{"className":6925,"style":6412},[47],[26,6927,6929,6932],{"className":6928},[47],[26,6930,251],{"className":6931},[47,179],[26,6933,6935],{"className":6934},[184],[26,6936,6938,6958],{"className":6937},[188,189],[26,6939,6941,6955],{"className":6940},[193],[26,6942,6944],{"className":6943,"style":264},[197],[26,6945,6946,6949],{"style":201},[26,6947],{"className":6948,"style":206},[205],[26,6950,6952],{"className":6951},[210,211,212,213],[26,6953,1282],{"className":6954},[47,213],[26,6956,221],{"className":6957},[220],[26,6959,6961],{"className":6960},[193],[26,6962,6964],{"className":6963,"style":228},[197],[26,6965],{},[26,6967,221],{"className":6968},[220],[26,6970,6972],{"className":6971},[193],[26,6973,6975],{"className":6974,"style":6526},[197],[26,6976],{},[26,6978],{"className":6979},[515,6385],[26,6981,524],{"className":6982},[523],[26,6984],{"className":6985,"style":3616},[54],[26,6987],{"className":6988,"style":392},[54],[26,6990,6992,6995,7138],{"className":6991},[47],[26,6993],{"className":6994},[400,6385],[26,6996,6998],{"className":6997},[6389],[26,6999,7001,7129],{"className":7000},[188,189],[26,7002,7004,7126],{"className":7003},[193],[26,7005,7007,7067,7075],{"className":7006,"style":6399},[197],[26,7008,7009,7012],{"style":6402},[26,7010],{"className":7011,"style":5134},[205],[26,7013,7015,7018],{"className":7014},[47],[26,7016,6413],{"className":7017,"style":6412},[47],[26,7019,7021,7024],{"className":7020},[47],[26,7022,251],{"className":7023},[47,179],[26,7025,7027],{"className":7026},[184],[26,7028,7030,7059],{"className":7029},[188,189],[26,7031,7033,7056],{"className":7032},[193],[26,7034,7036],{"className":7035,"style":264},[197],[26,7037,7038,7041],{"style":201},[26,7039],{"className":7040,"style":206},[205],[26,7042,7044],{"className":7043},[210,211,212,213],[26,7045,7047,7050,7053],{"className":7046},[47,213],[26,7048,158],{"className":7049},[47,48,213],[26,7051,60],{"className":7052},[59,213],[26,7054,74],{"className":7055},[47,213],[26,7057,221],{"className":7058},[220],[26,7060,7062],{"className":7061},[193],[26,7063,7065],{"className":7064,"style":294},[197],[26,7066],{},[26,7068,7069,7072],{"style":6456},[26,7070],{"className":7071,"style":5134},[205],[26,7073],{"className":7074,"style":6464},[6463],[26,7076,7077,7080],{"style":6467},[26,7078],{"className":7079,"style":5134},[205],[26,7081,7083,7086],{"className":7082},[47],[26,7084,6413],{"className":7085,"style":6412},[47],[26,7087,7089,7092],{"className":7088},[47],[26,7090,251],{"className":7091},[47,179],[26,7093,7095],{"className":7094},[184],[26,7096,7098,7118],{"className":7097},[188,189],[26,7099,7101,7115],{"className":7100},[193],[26,7102,7104],{"className":7103,"style":198},[197],[26,7105,7106,7109],{"style":201},[26,7107],{"className":7108,"style":206},[205],[26,7110,7112],{"className":7111},[210,211,212,213],[26,7113,158],{"className":7114},[47,48,213],[26,7116,221],{"className":7117},[220],[26,7119,7121],{"className":7120},[193],[26,7122,7124],{"className":7123,"style":228},[197],[26,7125],{},[26,7127,221],{"className":7128},[220],[26,7130,7132],{"className":7131},[193],[26,7133,7136],{"className":7134,"style":7135},[197],"height:0.8943em;",[26,7137],{},[26,7139],{"className":7140},[515,6385],[26,7142],{"className":7143,"style":358},[54],[26,7145,366],{"className":7146},[365],[26,7148],{"className":7149,"style":358},[54],[26,7151,7153,7157,7164,7167,7170,7261,7264,7267],{"className":7152},[38],[26,7154],{"className":7155,"style":7156},[42],"height:1.0519em;vertical-align:-0.25em;",[26,7158,7160],{"className":7159},[47],[26,7161,7163],{"className":7162,"style":5281},[47,698],"diag",[26,7165],{"className":7166,"style":388},[54],[26,7168],{"className":7169,"style":392},[54],[26,7171,7173,7176,7211,7214,7255,7258],{"className":7172},[396],[26,7174,403],{"className":7175,"style":402},[400,401],[26,7177,7179,7182],{"className":7178},[47],[26,7180,384],{"className":7181,"style":383},[47,48],[26,7183,7185],{"className":7184},[184],[26,7186,7188],{"className":7187},[188],[26,7189,7191],{"className":7190},[193],[26,7192,7195],{"className":7193,"style":7194},[197],"height:0.8019em;",[26,7196,7198,7201],{"style":7197},"top:-3.113em;margin-right:0.05em;",[26,7199],{"className":7200,"style":206},[205],[26,7202,7204],{"className":7203},[210,211,212,213],[26,7205,7207],{"className":7206},[47,213],[26,7208,7210],{"className":7209},[47,213],"′",[26,7212,403],{"className":7213},[400],[26,7215,7217,7221],{"className":7216},[47],[26,7218,7220],{"className":7219},[47,179],"z",[26,7222,7224],{"className":7223},[184],[26,7225,7227,7247],{"className":7226},[188,189],[26,7228,7230,7244],{"className":7229},[193],[26,7231,7233],{"className":7232,"style":198},[197],[26,7234,7235,7238],{"style":201},[26,7236],{"className":7237,"style":206},[205],[26,7239,7241],{"className":7240},[210,211,212,213],[26,7242,158],{"className":7243},[47,48,213],[26,7245,221],{"className":7246},[220],[26,7248,7250],{"className":7249},[193],[26,7251,7253],{"className":7252,"style":228},[197],[26,7254],{},[26,7256,516],{"className":7257},[515],[26,7259,516],{"className":7260,"style":402},[515,401],[26,7262],{"className":7263,"style":392},[54],[26,7265,407],{"className":7266},[47,179],[26,7268,524],{"className":7269},[523],[11,7271,527,7272,7449,7450,7453,7454,7515,7516,7533,7534,7549,7550,7566,7567,7582,7583,7598,7599,7615,7616,7598,7682,7698,7699,7765,7766,7615,7782,7848],{},[26,7273,7275],{"className":7274},[29],[26,7276,7278,7333,7400],{"className":7277,"ariaHidden":34},[33],[26,7279,7281,7284,7324,7327,7330],{"className":7280},[38],[26,7282],{"className":7283,"style":172},[42],[26,7285,7287,7290],{"className":7286},[47],[26,7288,7220],{"className":7289},[47,179],[26,7291,7293],{"className":7292},[184],[26,7294,7296,7316],{"className":7295},[188,189],[26,7297,7299,7313],{"className":7298},[193],[26,7300,7302],{"className":7301,"style":198},[197],[26,7303,7304,7307],{"style":201},[26,7305],{"className":7306,"style":206},[205],[26,7308,7310],{"className":7309},[210,211,212,213],[26,7311,158],{"className":7312},[47,48,213],[26,7314,221],{"className":7315},[220],[26,7317,7319],{"className":7318},[193],[26,7320,7322],{"className":7321,"style":228},[197],[26,7323],{},[26,7325],{"className":7326,"style":358},[54],[26,7328,366],{"className":7329},[365],[26,7331],{"className":7332,"style":358},[54],[26,7334,7336,7339,7342,7391,7394,7397],{"className":7335},[38],[26,7337],{"className":7338,"style":244},[42],[26,7340,407],{"className":7341},[47,179],[26,7343,7345,7348],{"className":7344},[47],[26,7346,251],{"className":7347},[47,179],[26,7349,7351],{"className":7350},[184],[26,7352,7354,7383],{"className":7353},[188,189],[26,7355,7357,7380],{"className":7356},[193],[26,7358,7360],{"className":7359,"style":264},[197],[26,7361,7362,7365],{"style":201},[26,7363],{"className":7364,"style":206},[205],[26,7366,7368],{"className":7367},[210,211,212,213],[26,7369,7371,7374,7377],{"className":7370},[47,213],[26,7372,158],{"className":7373},[47,48,213],[26,7375,60],{"className":7376},[59,213],[26,7378,74],{"className":7379},[47,213],[26,7381,221],{"className":7382},[220],[26,7384,7386],{"className":7385},[193],[26,7387,7389],{"className":7388,"style":294},[197],[26,7390],{},[26,7392],{"className":7393,"style":55},[54],[26,7395,463],{"className":7396},[59],[26,7398],{"className":7399,"style":55},[54],[26,7401,7403,7406,7409],{"className":7402},[38],[26,7404],{"className":7405,"style":2545},[42],[26,7407,471],{"className":7408,"style":470},[47,179],[26,7410,7412,7415],{"className":7411},[47],[26,7413,180],{"className":7414},[47,179],[26,7416,7418],{"className":7417},[184],[26,7419,7421,7441],{"className":7420},[188,189],[26,7422,7424,7438],{"className":7423},[193],[26,7425,7427],{"className":7426,"style":198},[197],[26,7428,7429,7432],{"style":201},[26,7430],{"className":7431,"style":206},[205],[26,7433,7435],{"className":7434},[210,211,212,213],[26,7436,158],{"className":7437},[47,48,213],[26,7439,221],{"className":7440},[220],[26,7442,7444],{"className":7443},[193],[26,7445,7447],{"className":7446,"style":228},[197],[26,7448],{},". Each hop back\nmultiplies by the ",[94,7451,7452],{},"same"," factor ",[26,7455,7457],{"className":7456},[29],[26,7458,7460],{"className":7459,"ariaHidden":34},[33],[26,7461,7463,7467,7473,7476,7509,7512],{"className":7462},[38],[26,7464],{"className":7465,"style":7466},[42],"height:1.0019em;vertical-align:-0.25em;",[26,7468,7470],{"className":7469},[47],[26,7471,7163],{"className":7472,"style":5281},[47,698],[26,7474,403],{"className":7475},[400],[26,7477,7479,7482],{"className":7478},[47],[26,7480,384],{"className":7481,"style":383},[47,48],[26,7483,7485],{"className":7484},[184],[26,7486,7488],{"className":7487},[188],[26,7489,7491],{"className":7490},[193],[26,7492,7495],{"className":7493,"style":7494},[197],"height:0.7519em;",[26,7496,7497,7500],{"style":824},[26,7498],{"className":7499,"style":206},[205],[26,7501,7503],{"className":7502},[210,211,212,213],[26,7504,7506],{"className":7505},[47,213],[26,7507,7210],{"className":7508},[47,213],[26,7510,516],{"className":7511},[515],[26,7513,407],{"className":7514},[47,179],", so crossing ",[26,7517,7519],{"className":7518},[29],[26,7520,7522],{"className":7521,"ariaHidden":34},[33],[26,7523,7525,7528],{"className":7524},[38],[26,7526],{"className":7527,"style":557},[42],[26,7529,7532],{"className":7530,"style":7531},[47,48],"margin-right:0.0315em;","k"," steps\nmultiplies by that factor ",[26,7535,7537],{"className":7536},[29],[26,7538,7540],{"className":7539,"ariaHidden":34},[33],[26,7541,7543,7546],{"className":7542},[38],[26,7544],{"className":7545,"style":557},[42],[26,7547,7532],{"className":7548,"style":7531},[47,48]," times. For example, suppose the recurrent\nweight and the local derivative combine to a factor of ",[26,7551,7553],{"className":7552},[29],[26,7554,7556],{"className":7555,"ariaHidden":34},[33],[26,7557,7559,7562],{"className":7558},[38],[26,7560],{"className":7561,"style":70},[42],[26,7563,7565],{"className":7564},[47],"0.6"," per step. Then a gradient\ncrossing ",[26,7568,7570],{"className":7569},[29],[26,7571,7573],{"className":7572,"ariaHidden":34},[33],[26,7574,7576,7579],{"className":7575},[38],[26,7577],{"className":7578,"style":70},[42],[26,7580,74],{"className":7581},[47]," step is scaled by ",[26,7584,7586],{"className":7585},[29],[26,7587,7589],{"className":7588,"ariaHidden":34},[33],[26,7590,7592,7595],{"className":7591},[38],[26,7593],{"className":7594,"style":70},[42],[26,7596,7565],{"className":7597},[47],", across ",[26,7600,7602],{"className":7601},[29],[26,7603,7605],{"className":7604,"ariaHidden":34},[33],[26,7606,7608,7611],{"className":7607},[38],[26,7609],{"className":7610,"style":70},[42],[26,7612,7614],{"className":7613},[47],"5"," steps by ",[26,7617,7619],{"className":7618},[29],[26,7620,7622,7672],{"className":7621,"ariaHidden":34},[33],[26,7623,7625,7629,7633,7663,7666,7669],{"className":7624},[38],[26,7626],{"className":7627,"style":7628},[42],"height:0.8141em;",[26,7630,7632],{"className":7631},[47],"0.",[26,7634,7636,7640],{"className":7635},[47],[26,7637,7639],{"className":7638},[47],"6",[26,7641,7643],{"className":7642},[184],[26,7644,7646],{"className":7645},[188],[26,7647,7649],{"className":7648},[193],[26,7650,7652],{"className":7651,"style":7628},[197],[26,7653,7654,7657],{"style":824},[26,7655],{"className":7656,"style":206},[205],[26,7658,7660],{"className":7659},[210,211,212,213],[26,7661,7614],{"className":7662},[47,213],[26,7664],{"className":7665,"style":358},[54],[26,7667,366],{"className":7668},[365],[26,7670],{"className":7671,"style":358},[54],[26,7673,7675,7678],{"className":7674},[38],[26,7676],{"className":7677,"style":70},[42],[26,7679,7681],{"className":7680},[47],"0.078",[26,7683,7685],{"className":7684},[29],[26,7686,7688],{"className":7687,"ariaHidden":34},[33],[26,7689,7691,7694],{"className":7690},[38],[26,7692],{"className":7693,"style":70},[42],[26,7695,7697],{"className":7696},[47],"10","\nsteps by ",[26,7700,7702],{"className":7701},[29],[26,7703,7705,7755],{"className":7704,"ariaHidden":34},[33],[26,7706,7708,7711,7714,7746,7749,7752],{"className":7707},[38],[26,7709],{"className":7710,"style":7628},[42],[26,7712,7632],{"className":7713},[47],[26,7715,7717,7720],{"className":7716},[47],[26,7718,7639],{"className":7719},[47],[26,7721,7723],{"className":7722},[184],[26,7724,7726],{"className":7725},[188],[26,7727,7729],{"className":7728},[193],[26,7730,7732],{"className":7731,"style":7628},[197],[26,7733,7734,7737],{"style":824},[26,7735],{"className":7736,"style":206},[205],[26,7738,7740],{"className":7739},[210,211,212,213],[26,7741,7743],{"className":7742},[47,213],[26,7744,7697],{"className":7745},[47,213],[26,7747],{"className":7748,"style":358},[54],[26,7750,366],{"className":7751},[365],[26,7753],{"className":7754,"style":358},[54],[26,7756,7758,7761],{"className":7757},[38],[26,7759],{"className":7760,"style":70},[42],[26,7762,7764],{"className":7763},[47],"0.006",", and across ",[26,7767,7769],{"className":7768},[29],[26,7770,7772],{"className":7771,"ariaHidden":34},[33],[26,7773,7775,7778],{"className":7774},[38],[26,7776],{"className":7777,"style":70},[42],[26,7779,7781],{"className":7780},[47],"20",[26,7783,7785],{"className":7784},[29],[26,7786,7788,7838],{"className":7787,"ariaHidden":34},[33],[26,7789,7791,7794,7797,7829,7832,7835],{"className":7790},[38],[26,7792],{"className":7793,"style":7628},[42],[26,7795,7632],{"className":7796},[47],[26,7798,7800,7803],{"className":7799},[47],[26,7801,7639],{"className":7802},[47],[26,7804,7806],{"className":7805},[184],[26,7807,7809],{"className":7808},[188],[26,7810,7812],{"className":7811},[193],[26,7813,7815],{"className":7814,"style":7628},[197],[26,7816,7817,7820],{"style":824},[26,7818],{"className":7819,"style":206},[205],[26,7821,7823],{"className":7822},[210,211,212,213],[26,7824,7826],{"className":7825},[47,213],[26,7827,7781],{"className":7828},[47,213],[26,7830],{"className":7831,"style":358},[54],[26,7833,366],{"className":7834},[365],[26,7836],{"className":7837,"style":358},[54],[26,7839,7841,7844],{"className":7840},[38],[26,7842],{"className":7843,"style":70},[42],[26,7845,7847],{"className":7846},[47],"0.00004",". The signal\nreaching a state twenty words back is four orders of magnitude weaker than the one\nreaching the immediately previous state.",[1319,7850],{"hash":7851},"98d0012c7dfa26b76a946676e25837d1c65d9435fc3dbc601a51455a14552017",[11,7853,7854,7855,7883],{},"The same product governs ",[26,7856,7858],{"className":7857},[29],[26,7859,7861],{"className":7860,"ariaHidden":34},[33],[26,7862,7864,7867,7870,7873,7877,7880],{"className":7863},[38],[26,7865],{"className":7866,"style":379},[42],[26,7868,6413],{"className":7869,"style":6412},[47],[26,7871,5065],{"className":7872},[47,48],[26,7874,7876],{"className":7875},[47],"\u002F",[26,7878,6413],{"className":7879,"style":6412},[47],[26,7881,407],{"className":7882},[47,179]," and the other weight\ngradients, so a plain RNN cannot learn to connect a decision to context more than a\nhandful of steps back — precisely the failure the next section names.",[127,7885,7887],{"id":7886},"the-vanishing-gradient-problem","The vanishing-gradient problem",[11,7889,7890,7891,7906,7907,7941,7942,7957,7958,7973,7974,7989,7990,8005],{},"Consider how an error at step ",[26,7892,7894],{"className":7893},[29],[26,7895,7897],{"className":7896,"ariaHidden":34},[33],[26,7898,7900,7903],{"className":7899},[38],[26,7901],{"className":7902,"style":154},[42],[26,7904,158],{"className":7905},[47,48]," reaches a hidden state many steps earlier, at step\n",[26,7908,7910],{"className":7909},[29],[26,7911,7913,7932],{"className":7912,"ariaHidden":34},[33],[26,7914,7916,7920,7923,7926,7929],{"className":7915},[38],[26,7917],{"className":7918,"style":7919},[42],"height:0.6984em;vertical-align:-0.0833em;",[26,7921,158],{"className":7922},[47,48],[26,7924],{"className":7925,"style":55},[54],[26,7927,60],{"className":7928},[59],[26,7930],{"className":7931,"style":55},[54],[26,7933,7935,7938],{"className":7934},[38],[26,7936],{"className":7937,"style":557},[42],[26,7939,7532],{"className":7940,"style":7531},[47,48],". The gradient must flow backward through the recurrence once per intervening step, and\neach such hop multiplies by the recurrent weight ",[26,7943,7945],{"className":7944},[29],[26,7946,7948],{"className":7947,"ariaHidden":34},[33],[26,7949,7951,7954],{"className":7950},[38],[26,7952],{"className":7953,"style":576},[42],[26,7955,407],{"className":7956},[47,179]," and by the derivative of the\nnon-linearity ",[26,7959,7961],{"className":7960},[29],[26,7962,7964],{"className":7963,"ariaHidden":34},[33],[26,7965,7967,7970],{"className":7966},[38],[26,7968],{"className":7969,"style":540},[42],[26,7971,384],{"className":7972,"style":383},[47,48],". Over ",[26,7975,7977],{"className":7976},[29],[26,7978,7980],{"className":7979,"ariaHidden":34},[33],[26,7981,7983,7986],{"className":7982},[38],[26,7984],{"className":7985,"style":557},[42],[26,7987,7532],{"className":7988,"style":7531},[47,48]," steps the signal is multiplied by roughly the same factor ",[26,7991,7993],{"className":7992},[29],[26,7994,7996],{"className":7995,"ariaHidden":34},[33],[26,7997,7999,8002],{"className":7998},[38],[26,8000],{"className":8001,"style":557},[42],[26,8003,7532],{"className":8004,"style":7531},[47,48],"\ntimes — a geometric progression.",[11,8007,8008,8009,8012,8013,8016,8017],{},"If that repeated factor is less than one, the product shrinks toward zero: the gradient\n",[21,8010,8011],{},"vanishes",", and the far-past hidden states receive essentially no learning signal. If the\nfactor exceeds one, the product blows up and the gradient ",[21,8014,8015],{},"explodes",". In practice, plain RNNs\nvanish far more often than they explode, and the effect is that the network cannot tie a\ndecision to context more than a handful of steps back.",[117,8018,8019],{},[15,8020,8024],{"href":8021,"ariaDescribedBy":8022,"dataFootnoteRef":6,"id":8023},"#user-content-fn-jm-vanish",[123],"user-content-fnref-jm-vanish","4",[1319,8026],{"hash":8027},"4af20d242d8806800bf5557f3d5439c8d73f20c1ed7434f8e303dc6830e8c45e",[11,8029,8030,8031,8034,8035,8038,8039,8046,8047,8049,8050,8053,8054,8057],{},"The deeper diagnosis is that a plain RNN requires one hidden layer to do two conflicting jobs at\nonce: supply the information the ",[94,8032,8033],{},"current"," step needs, and preserve information that some\n",[94,8036,8037],{},"future"," step will need. Every write for the present overwrites part of the past. Consider\n",[94,8040,8041,8042,8045],{},"the flights the airline was cancelling ",[21,8043,8044],{},"were"," full",": the number of ",[94,8048,8044],{}," depends on\n",[94,8051,8052],{},"flights",", but the intervening singular ",[94,8055,8056],{},"airline"," dominates the local context and the plural\nsignal is written over long before it is needed. A network with a single, always-updated state\nhas no way to hold a value aside untouched.",[127,8059,8061],{"id":8060},"the-lstm","The LSTM",[11,8063,13,8064,8067,8068,3257,8071,8124,8125,8132,8133,8136],{},[21,8065,8066],{},"long short-term memory"," network solves this by giving the unit a second state — a\n",[21,8069,8070],{},"cell state",[26,8072,8074],{"className":8073},[29],[26,8075,8077],{"className":8076,"ariaHidden":34},[33],[26,8078,8080,8083],{"className":8079},[38],[26,8081],{"className":8082,"style":172},[42],[26,8084,8086,8090],{"className":8085},[47],[26,8087,8089],{"className":8088},[47,179],"c",[26,8091,8093],{"className":8092},[184],[26,8094,8096,8116],{"className":8095},[188,189],[26,8097,8099,8113],{"className":8098},[193],[26,8100,8102],{"className":8101,"style":198},[197],[26,8103,8104,8107],{"style":201},[26,8105],{"className":8106,"style":206},[205],[26,8108,8110],{"className":8109},[210,211,212,213],[26,8111,158],{"className":8112},[47,48,213],[26,8114,221],{"className":8115},[220],[26,8117,8119],{"className":8118},[193],[26,8120,8122],{"className":8121,"style":228},[197],[26,8123],{}," that runs alongside the hidden state and is engineered so that\ninformation can pass along it nearly unchanged.",[117,8126,8127],{},[15,8128,7614],{"href":8129,"ariaDescribedBy":8130,"dataFootnoteRef":6,"id":8131},"#user-content-fn-jm-lstm",[123],"user-content-fnref-jm-lstm"," The network then learns, through small\nsub-networks called ",[21,8134,8135],{},"gates",", which information to erase from the cell, which to add, and which\nto expose as output. Context management is learned rather than fixed.",[11,8138,8139,8140,8157,8158,8188,8189,8206,8207,8222,8223,8238,8239,8254,8255,8270],{},"Every gate shares one design: a feedforward layer, a sigmoid ",[26,8141,8143],{"className":8142},[29],[26,8144,8146],{"className":8145,"ariaHidden":34},[33],[26,8147,8149,8153],{"className":8148},[38],[26,8150],{"className":8151,"style":8152},[42],"height:0.4306em;",[26,8154,8156],{"className":8155,"style":383},[47,48],"σ"," squashing its output into\n",[26,8159,8161],{"className":8160},[29],[26,8162,8164],{"className":8163,"ariaHidden":34},[33],[26,8165,8167,8170,8173,8176,8179,8182,8185],{"className":8166},[38],[26,8168],{"className":8169,"style":379},[42],[26,8171,4301],{"className":8172},[400],[26,8174,2291],{"className":8175},[47],[26,8177,524],{"className":8178},[523],[26,8180],{"className":8181,"style":392},[54],[26,8183,74],{"className":8184},[47],[26,8186,4308],{"className":8187},[515],", and a pointwise (Hadamard) multiplication ",[26,8190,8192],{"className":8191},[29],[26,8193,8195],{"className":8194,"ariaHidden":34},[33],[26,8196,8198,8202],{"className":8197},[38],[26,8199],{"className":8200,"style":8201},[42],"height:0.6667em;vertical-align:-0.0833em;",[26,8203,8205],{"className":8204},[47],"⊙"," with the vector being gated. Because\nthe sigmoid pushes toward ",[26,8208,8210],{"className":8209},[29],[26,8211,8213],{"className":8212,"ariaHidden":34},[33],[26,8214,8216,8219],{"className":8215},[38],[26,8217],{"className":8218,"style":70},[42],[26,8220,2291],{"className":8221},[47]," or ",[26,8224,8226],{"className":8225},[29],[26,8227,8229],{"className":8228,"ariaHidden":34},[33],[26,8230,8232,8235],{"className":8231},[38],[26,8233],{"className":8234,"style":70},[42],[26,8236,74],{"className":8237},[47],", the gate acts as a soft binary mask — coordinates near ",[26,8240,8242],{"className":8241},[29],[26,8243,8245],{"className":8244,"ariaHidden":34},[33],[26,8246,8248,8251],{"className":8247},[38],[26,8249],{"className":8250,"style":70},[42],[26,8252,74],{"className":8253},[47],"\npass through, coordinates near ",[26,8256,8258],{"className":8257},[29],[26,8259,8261],{"className":8260,"ariaHidden":34},[33],[26,8262,8264,8267],{"className":8263},[38],[26,8265],{"className":8266,"style":70},[42],[26,8268,2291],{"className":8269},[47]," are erased.",[11,8272,8273,8276,8277,8331,8332,8394],{},[21,8274,8275],{},"The forget gate"," decides what to delete from the previous cell state. It reads the previous\nhidden state and current input, and its mask ",[26,8278,8280],{"className":8279},[29],[26,8281,8283],{"className":8282,"ariaHidden":34},[33],[26,8284,8286,8289],{"className":8285},[38],[26,8287],{"className":8288,"style":314},[42],[26,8290,8292,8296],{"className":8291},[47],[26,8293,8295],{"className":8294,"style":49},[47,179],"f",[26,8297,8299],{"className":8298},[184],[26,8300,8302,8323],{"className":8301},[188,189],[26,8303,8305,8320],{"className":8304},[193],[26,8306,8308],{"className":8307,"style":198},[197],[26,8309,8311,8314],{"style":8310},"top:-2.55em;margin-left:-0.109em;margin-right:0.05em;",[26,8312],{"className":8313,"style":206},[205],[26,8315,8317],{"className":8316},[210,211,212,213],[26,8318,158],{"className":8319},[47,48,213],[26,8321,221],{"className":8322},[220],[26,8324,8326],{"className":8325},[193],[26,8327,8329],{"className":8328,"style":228},[197],[26,8330],{}," multiplies ",[26,8333,8335],{"className":8334},[29],[26,8336,8338],{"className":8337,"ariaHidden":34},[33],[26,8339,8341,8345],{"className":8340},[38],[26,8342],{"className":8343,"style":8344},[42],"height:0.6528em;vertical-align:-0.2083em;",[26,8346,8348,8351],{"className":8347},[47],[26,8349,8089],{"className":8350},[47,179],[26,8352,8354],{"className":8353},[184],[26,8355,8357,8386],{"className":8356},[188,189],[26,8358,8360,8383],{"className":8359},[193],[26,8361,8363],{"className":8362,"style":264},[197],[26,8364,8365,8368],{"style":201},[26,8366],{"className":8367,"style":206},[205],[26,8369,8371],{"className":8370},[210,211,212,213],[26,8372,8374,8377,8380],{"className":8373},[47,213],[26,8375,158],{"className":8376},[47,48,213],[26,8378,60],{"className":8379},[59,213],[26,8381,74],{"className":8382},[47,213],[26,8384,221],{"className":8385},[220],[26,8387,8389],{"className":8388},[193],[26,8390,8392],{"className":8391,"style":294},[197],[26,8393],{},"\nelement-wise, zeroing out whatever is no longer relevant:",[26,8396,8398],{"className":8397},[301],[26,8399,8401],{"className":8400},[29],[26,8402,8404,8465,8738,8803],{"className":8403,"ariaHidden":34},[33],[26,8405,8407,8410,8450,8453,8456,8459,8462],{"className":8406},[38],[26,8408],{"className":8409,"style":314},[42],[26,8411,8413,8416],{"className":8412},[47],[26,8414,8295],{"className":8415,"style":49},[47,179],[26,8417,8419],{"className":8418},[184],[26,8420,8422,8442],{"className":8421},[188,189],[26,8423,8425,8439],{"className":8424},[193],[26,8426,8428],{"className":8427,"style":198},[197],[26,8429,8430,8433],{"style":8310},[26,8431],{"className":8432,"style":206},[205],[26,8434,8436],{"className":8435},[210,211,212,213],[26,8437,158],{"className":8438},[47,48,213],[26,8440,221],{"className":8441},[220],[26,8443,8445],{"className":8444},[193],[26,8446,8448],{"className":8447,"style":228},[197],[26,8449],{},[26,8451],{"className":8452,"style":358},[54],[26,8454],{"className":8455,"style":358},[54],[26,8457,366],{"className":8458},[365],[26,8460],{"className":8461,"style":358},[54],[26,8463],{"className":8464,"style":358},[54],[26,8466,8468,8472,8475,8478,8481,8671,8674,8677,8680,8683,8723,8726,8729,8732,8735],{"className":8467},[38],[26,8469],{"className":8470,"style":8471},[42],"height:1.0361em;vertical-align:-0.2861em;",[26,8473,8156],{"className":8474,"style":383},[47,48],[26,8476],{"className":8477,"style":388},[54],[26,8479],{"className":8480,"style":392},[54],[26,8482,8484,8487,8530,8579,8582,8585,8588,8628,8668],{"className":8483},[396],[26,8485,403],{"className":8486,"style":402},[400,401],[26,8488,8490,8493],{"className":8489},[47],[26,8491,407],{"className":8492},[47,179],[26,8494,8496],{"className":8495},[184],[26,8497,8499,8521],{"className":8498},[188,189],[26,8500,8502,8518],{"className":8501},[193],[26,8503,8506],{"className":8504,"style":8505},[197],"height:0.3361em;",[26,8507,8508,8511],{"style":201},[26,8509],{"className":8510,"style":206},[205],[26,8512,8514],{"className":8513},[210,211,212,213],[26,8515,8295],{"className":8516,"style":8517},[47,48,213],"margin-right:0.1076em;",[26,8519,221],{"className":8520},[220],[26,8522,8524],{"className":8523},[193],[26,8525,8528],{"className":8526,"style":8527},[197],"height:0.2861em;",[26,8529],{},[26,8531,8533,8536],{"className":8532},[47],[26,8534,251],{"className":8535},[47,179],[26,8537,8539],{"className":8538},[184],[26,8540,8542,8571],{"className":8541},[188,189],[26,8543,8545,8568],{"className":8544},[193],[26,8546,8548],{"className":8547,"style":264},[197],[26,8549,8550,8553],{"style":201},[26,8551],{"className":8552,"style":206},[205],[26,8554,8556],{"className":8555},[210,211,212,213],[26,8557,8559,8562,8565],{"className":8558},[47,213],[26,8560,158],{"className":8561},[47,48,213],[26,8563,60],{"className":8564},[59,213],[26,8566,74],{"className":8567},[47,213],[26,8569,221],{"className":8570},[220],[26,8572,8574],{"className":8573},[193],[26,8575,8577],{"className":8576,"style":294},[197],[26,8578],{},[26,8580],{"className":8581,"style":55},[54],[26,8583,463],{"className":8584},[59],[26,8586],{"className":8587,"style":55},[54],[26,8589,8591,8594],{"className":8590},[47],[26,8592,471],{"className":8593,"style":470},[47,179],[26,8595,8597],{"className":8596},[184],[26,8598,8600,8620],{"className":8599},[188,189],[26,8601,8603,8617],{"className":8602},[193],[26,8604,8606],{"className":8605,"style":8505},[197],[26,8607,8608,8611],{"style":650},[26,8609],{"className":8610,"style":206},[205],[26,8612,8614],{"className":8613},[210,211,212,213],[26,8615,8295],{"className":8616,"style":8517},[47,48,213],[26,8618,221],{"className":8619},[220],[26,8621,8623],{"className":8622},[193],[26,8624,8626],{"className":8625,"style":8527},[197],[26,8627],{},[26,8629,8631,8634],{"className":8630},[47],[26,8632,180],{"className":8633},[47,179],[26,8635,8637],{"className":8636},[184],[26,8638,8640,8660],{"className":8639},[188,189],[26,8641,8643,8657],{"className":8642},[193],[26,8644,8646],{"className":8645,"style":198},[197],[26,8647,8648,8651],{"style":201},[26,8649],{"className":8650,"style":206},[205],[26,8652,8654],{"className":8653},[210,211,212,213],[26,8655,158],{"className":8656},[47,48,213],[26,8658,221],{"className":8659},[220],[26,8661,8663],{"className":8662},[193],[26,8664,8666],{"className":8665,"style":228},[197],[26,8667],{},[26,8669,516],{"className":8670,"style":402},[515,401],[26,8672],{"className":8673,"style":392},[54],[26,8675,524],{"className":8676},[523],[26,8678],{"className":8679,"style":3616},[54],[26,8681],{"className":8682,"style":392},[54],[26,8684,8686,8689],{"className":8685},[47],[26,8687,7532],{"className":8688},[47,179],[26,8690,8692],{"className":8691},[184],[26,8693,8695,8715],{"className":8694},[188,189],[26,8696,8698,8712],{"className":8697},[193],[26,8699,8701],{"className":8700,"style":198},[197],[26,8702,8703,8706],{"style":201},[26,8704],{"className":8705,"style":206},[205],[26,8707,8709],{"className":8708},[210,211,212,213],[26,8710,158],{"className":8711},[47,48,213],[26,8713,221],{"className":8714},[220],[26,8716,8718],{"className":8717},[193],[26,8719,8721],{"className":8720,"style":228},[197],[26,8722],{},[26,8724],{"className":8725,"style":358},[54],[26,8727],{"className":8728,"style":358},[54],[26,8730,366],{"className":8731},[365],[26,8733],{"className":8734,"style":358},[54],[26,8736],{"className":8737,"style":358},[54],[26,8739,8741,8745,8794,8797,8800],{"className":8740},[38],[26,8742],{"className":8743,"style":8744},[42],"height:0.7917em;vertical-align:-0.2083em;",[26,8746,8748,8751],{"className":8747},[47],[26,8749,8089],{"className":8750},[47,179],[26,8752,8754],{"className":8753},[184],[26,8755,8757,8786],{"className":8756},[188,189],[26,8758,8760,8783],{"className":8759},[193],[26,8761,8763],{"className":8762,"style":264},[197],[26,8764,8765,8768],{"style":201},[26,8766],{"className":8767,"style":206},[205],[26,8769,8771],{"className":8770},[210,211,212,213],[26,8772,8774,8777,8780],{"className":8773},[47,213],[26,8775,158],{"className":8776},[47,48,213],[26,8778,60],{"className":8779},[59,213],[26,8781,74],{"className":8782},[47,213],[26,8784,221],{"className":8785},[220],[26,8787,8789],{"className":8788},[193],[26,8790,8792],{"className":8791,"style":294},[197],[26,8793],{},[26,8795],{"className":8796,"style":55},[54],[26,8798,8205],{"className":8799},[59],[26,8801],{"className":8802,"style":55},[54],[26,8804,8806,8809,8849],{"className":8805},[38],[26,8807],{"className":8808,"style":314},[42],[26,8810,8812,8815],{"className":8811},[47],[26,8813,8295],{"className":8814,"style":49},[47,179],[26,8816,8818],{"className":8817},[184],[26,8819,8821,8841],{"className":8820},[188,189],[26,8822,8824,8838],{"className":8823},[193],[26,8825,8827],{"className":8826,"style":198},[197],[26,8828,8829,8832],{"style":8310},[26,8830],{"className":8831,"style":206},[205],[26,8833,8835],{"className":8834},[210,211,212,213],[26,8836,158],{"className":8837},[47,48,213],[26,8839,221],{"className":8840},[220],[26,8842,8844],{"className":8843},[193],[26,8845,8847],{"className":8846,"style":228},[197],[26,8848],{},[26,8850,765],{"className":8851},[47],[11,8853,8854,8857,8858,8873],{},[21,8855,8856],{},"The candidate"," is the new information proposed for the cell — the same ",[26,8859,8861],{"className":8860},[29],[26,8862,8864],{"className":8863,"ariaHidden":34},[33],[26,8865,8867,8870],{"className":8866},[38],[26,8868],{"className":8869,"style":557},[42],[26,8871,562],{"className":8872},[561]," computation a\nplain RNN would perform:",[26,8875,8877],{"className":8876},[301],[26,8878,8880],{"className":8879},[29],[26,8881,8883,8944],{"className":8882,"ariaHidden":34},[33],[26,8884,8886,8889,8929,8932,8935,8938,8941],{"className":8885},[38],[26,8887],{"className":8888,"style":628},[42],[26,8890,8892,8895],{"className":8891},[47],[26,8893,384],{"className":8894,"style":470},[47,179],[26,8896,8898],{"className":8897},[184],[26,8899,8901,8921],{"className":8900},[188,189],[26,8902,8904,8918],{"className":8903},[193],[26,8905,8907],{"className":8906,"style":198},[197],[26,8908,8909,8912],{"style":650},[26,8910],{"className":8911,"style":206},[205],[26,8913,8915],{"className":8914},[210,211,212,213],[26,8916,158],{"className":8917},[47,48,213],[26,8919,221],{"className":8920},[220],[26,8922,8924],{"className":8923},[193],[26,8925,8927],{"className":8926,"style":228},[197],[26,8928],{},[26,8930],{"className":8931,"style":358},[54],[26,8933],{"className":8934,"style":358},[54],[26,8936,366],{"className":8937},[365],[26,8939],{"className":8940,"style":358},[54],[26,8942],{"className":8943,"style":358},[54],[26,8945,8947,8950,8953,8956,8959,9146,9149],{"className":8946},[38],[26,8948],{"className":8949,"style":8471},[42],[26,8951,562],{"className":8952},[561],[26,8954],{"className":8955,"style":388},[54],[26,8957],{"className":8958,"style":392},[54],[26,8960,8962,8965,9005,9054,9057,9060,9063,9103,9143],{"className":8961},[396],[26,8963,403],{"className":8964,"style":402},[400,401],[26,8966,8968,8971],{"className":8967},[47],[26,8969,407],{"className":8970},[47,179],[26,8972,8974],{"className":8973},[184],[26,8975,8977,8997],{"className":8976},[188,189],[26,8978,8980,8994],{"className":8979},[193],[26,8981,8983],{"className":8982,"style":1539},[197],[26,8984,8985,8988],{"style":201},[26,8986],{"className":8987,"style":206},[205],[26,8989,8991],{"className":8990},[210,211,212,213],[26,8992,384],{"className":8993,"style":383},[47,48,213],[26,8995,221],{"className":8996},[220],[26,8998,9000],{"className":8999},[193],[26,9001,9003],{"className":9002,"style":8527},[197],[26,9004],{},[26,9006,9008,9011],{"className":9007},[47],[26,9009,251],{"className":9010},[47,179],[26,9012,9014],{"className":9013},[184],[26,9015,9017,9046],{"className":9016},[188,189],[26,9018,9020,9043],{"className":9019},[193],[26,9021,9023],{"className":9022,"style":264},[197],[26,9024,9025,9028],{"style":201},[26,9026],{"className":9027,"style":206},[205],[26,9029,9031],{"className":9030},[210,211,212,213],[26,9032,9034,9037,9040],{"className":9033},[47,213],[26,9035,158],{"className":9036},[47,48,213],[26,9038,60],{"className":9039},[59,213],[26,9041,74],{"className":9042},[47,213],[26,9044,221],{"className":9045},[220],[26,9047,9049],{"className":9048},[193],[26,9050,9052],{"className":9051,"style":294},[197],[26,9053],{},[26,9055],{"className":9056,"style":55},[54],[26,9058,463],{"className":9059},[59],[26,9061],{"className":9062,"style":55},[54],[26,9064,9066,9069],{"className":9065},[47],[26,9067,471],{"className":9068,"style":470},[47,179],[26,9070,9072],{"className":9071},[184],[26,9073,9075,9095],{"className":9074},[188,189],[26,9076,9078,9092],{"className":9077},[193],[26,9079,9081],{"className":9080,"style":1539},[197],[26,9082,9083,9086],{"style":650},[26,9084],{"className":9085,"style":206},[205],[26,9087,9089],{"className":9088},[210,211,212,213],[26,9090,384],{"className":9091,"style":383},[47,48,213],[26,9093,221],{"className":9094},[220],[26,9096,9098],{"className":9097},[193],[26,9099,9101],{"className":9100,"style":8527},[197],[26,9102],{},[26,9104,9106,9109],{"className":9105},[47],[26,9107,180],{"className":9108},[47,179],[26,9110,9112],{"className":9111},[184],[26,9113,9115,9135],{"className":9114},[188,189],[26,9116,9118,9132],{"className":9117},[193],[26,9119,9121],{"className":9120,"style":198},[197],[26,9122,9123,9126],{"style":201},[26,9124],{"className":9125,"style":206},[205],[26,9127,9129],{"className":9128},[210,211,212,213],[26,9130,158],{"className":9131},[47,48,213],[26,9133,221],{"className":9134},[220],[26,9136,9138],{"className":9137},[193],[26,9139,9141],{"className":9140,"style":228},[197],[26,9142],{},[26,9144,516],{"className":9145,"style":402},[515,401],[26,9147],{"className":9148,"style":392},[54],[26,9150,765],{"className":9151},[47],[11,9153,9154,9157],{},[21,9155,9156],{},"The input gate"," selects which of that candidate to actually add, then the surviving old\ncontext and the selected new context are summed into the updated cell state:",[26,9159,9161],{"className":9160},[301],[26,9162,9164],{"className":9163},[29],[26,9165,9167,9228,9497,9552,9608],{"className":9166,"ariaHidden":34},[33],[26,9168,9170,9173,9213,9216,9219,9222,9225],{"className":9169},[38],[26,9171],{"className":9172,"style":314},[42],[26,9174,9176,9179],{"className":9175},[47],[26,9177,4018],{"className":9178},[47,179],[26,9180,9182],{"className":9181},[184],[26,9183,9185,9205],{"className":9184},[188,189],[26,9186,9188,9202],{"className":9187},[193],[26,9189,9191],{"className":9190,"style":198},[197],[26,9192,9193,9196],{"style":201},[26,9194],{"className":9195,"style":206},[205],[26,9197,9199],{"className":9198},[210,211,212,213],[26,9200,158],{"className":9201},[47,48,213],[26,9203,221],{"className":9204},[220],[26,9206,9208],{"className":9207},[193],[26,9209,9211],{"className":9210,"style":228},[197],[26,9212],{},[26,9214],{"className":9215,"style":358},[54],[26,9217],{"className":9218,"style":358},[54],[26,9220,366],{"className":9221},[365],[26,9223],{"className":9224,"style":358},[54],[26,9226],{"className":9227,"style":358},[54],[26,9229,9231,9234,9237,9240,9243,9430,9433,9436,9439,9442,9482,9485,9488,9491,9494],{"className":9230},[38],[26,9232],{"className":9233,"style":379},[42],[26,9235,8156],{"className":9236,"style":383},[47,48],[26,9238],{"className":9239,"style":388},[54],[26,9241],{"className":9242,"style":392},[54],[26,9244,9246,9249,9289,9338,9341,9344,9347,9387,9427],{"className":9245},[396],[26,9247,403],{"className":9248,"style":402},[400,401],[26,9250,9252,9255],{"className":9251},[47],[26,9253,407],{"className":9254},[47,179],[26,9256,9258],{"className":9257},[184],[26,9259,9261,9281],{"className":9260},[188,189],[26,9262,9264,9278],{"className":9263},[193],[26,9265,9267],{"className":9266,"style":4510},[197],[26,9268,9269,9272],{"style":201},[26,9270],{"className":9271,"style":206},[205],[26,9273,9275],{"className":9274},[210,211,212,213],[26,9276,4018],{"className":9277},[47,48,213],[26,9279,221],{"className":9280},[220],[26,9282,9284],{"className":9283},[193],[26,9285,9287],{"className":9286,"style":228},[197],[26,9288],{},[26,9290,9292,9295],{"className":9291},[47],[26,9293,251],{"className":9294},[47,179],[26,9296,9298],{"className":9297},[184],[26,9299,9301,9330],{"className":9300},[188,189],[26,9302,9304,9327],{"className":9303},[193],[26,9305,9307],{"className":9306,"style":264},[197],[26,9308,9309,9312],{"style":201},[26,9310],{"className":9311,"style":206},[205],[26,9313,9315],{"className":9314},[210,211,212,213],[26,9316,9318,9321,9324],{"className":9317},[47,213],[26,9319,158],{"className":9320},[47,48,213],[26,9322,60],{"className":9323},[59,213],[26,9325,74],{"className":9326},[47,213],[26,9328,221],{"className":9329},[220],[26,9331,9333],{"className":9332},[193],[26,9334,9336],{"className":9335,"style":294},[197],[26,9337],{},[26,9339],{"className":9340,"style":55},[54],[26,9342,463],{"className":9343},[59],[26,9345],{"className":9346,"style":55},[54],[26,9348,9350,9353],{"className":9349},[47],[26,9351,471],{"className":9352,"style":470},[47,179],[26,9354,9356],{"className":9355},[184],[26,9357,9359,9379],{"className":9358},[188,189],[26,9360,9362,9376],{"className":9361},[193],[26,9363,9365],{"className":9364,"style":4510},[197],[26,9366,9367,9370],{"style":650},[26,9368],{"className":9369,"style":206},[205],[26,9371,9373],{"className":9372},[210,211,212,213],[26,9374,4018],{"className":9375},[47,48,213],[26,9377,221],{"className":9378},[220],[26,9380,9382],{"className":9381},[193],[26,9383,9385],{"className":9384,"style":228},[197],[26,9386],{},[26,9388,9390,9393],{"className":9389},[47],[26,9391,180],{"className":9392},[47,179],[26,9394,9396],{"className":9395},[184],[26,9397,9399,9419],{"className":9398},[188,189],[26,9400,9402,9416],{"className":9401},[193],[26,9403,9405],{"className":9404,"style":198},[197],[26,9406,9407,9410],{"style":201},[26,9408],{"className":9409,"style":206},[205],[26,9411,9413],{"className":9412},[210,211,212,213],[26,9414,158],{"className":9415},[47,48,213],[26,9417,221],{"className":9418},[220],[26,9420,9422],{"className":9421},[193],[26,9423,9425],{"className":9424,"style":228},[197],[26,9426],{},[26,9428,516],{"className":9429,"style":402},[515,401],[26,9431],{"className":9432,"style":392},[54],[26,9434,524],{"className":9435},[523],[26,9437],{"className":9438,"style":3616},[54],[26,9440],{"className":9441,"style":392},[54],[26,9443,9445,9448],{"className":9444},[47],[26,9446,8089],{"className":9447},[47,179],[26,9449,9451],{"className":9450},[184],[26,9452,9454,9474],{"className":9453},[188,189],[26,9455,9457,9471],{"className":9456},[193],[26,9458,9460],{"className":9459,"style":198},[197],[26,9461,9462,9465],{"style":201},[26,9463],{"className":9464,"style":206},[205],[26,9466,9468],{"className":9467},[210,211,212,213],[26,9469,158],{"className":9470},[47,48,213],[26,9472,221],{"className":9473},[220],[26,9475,9477],{"className":9476},[193],[26,9478,9480],{"className":9479,"style":228},[197],[26,9481],{},[26,9483],{"className":9484,"style":358},[54],[26,9486],{"className":9487,"style":358},[54],[26,9489,366],{"className":9490},[365],[26,9492],{"className":9493,"style":358},[54],[26,9495],{"className":9496,"style":358},[54],[26,9498,9500,9503,9543,9546,9549],{"className":9499},[38],[26,9501],{"className":9502,"style":314},[42],[26,9504,9506,9509],{"className":9505},[47],[26,9507,7532],{"className":9508},[47,179],[26,9510,9512],{"className":9511},[184],[26,9513,9515,9535],{"className":9514},[188,189],[26,9516,9518,9532],{"className":9517},[193],[26,9519,9521],{"className":9520,"style":198},[197],[26,9522,9523,9526],{"style":201},[26,9524],{"className":9525,"style":206},[205],[26,9527,9529],{"className":9528},[210,211,212,213],[26,9530,158],{"className":9531},[47,48,213],[26,9533,221],{"className":9534},[220],[26,9536,9538],{"className":9537},[193],[26,9539,9541],{"className":9540,"style":228},[197],[26,9542],{},[26,9544],{"className":9545,"style":55},[54],[26,9547,463],{"className":9548},[59],[26,9550],{"className":9551,"style":55},[54],[26,9553,9555,9559,9599,9602,9605],{"className":9554},[38],[26,9556],{"className":9557,"style":9558},[42],"height:0.7778em;vertical-align:-0.1944em;",[26,9560,9562,9565],{"className":9561},[47],[26,9563,384],{"className":9564,"style":470},[47,179],[26,9566,9568],{"className":9567},[184],[26,9569,9571,9591],{"className":9570},[188,189],[26,9572,9574,9588],{"className":9573},[193],[26,9575,9577],{"className":9576,"style":198},[197],[26,9578,9579,9582],{"style":650},[26,9580],{"className":9581,"style":206},[205],[26,9583,9585],{"className":9584},[210,211,212,213],[26,9586,158],{"className":9587},[47,48,213],[26,9589,221],{"className":9590},[220],[26,9592,9594],{"className":9593},[193],[26,9595,9597],{"className":9596,"style":228},[197],[26,9598],{},[26,9600],{"className":9601,"style":55},[54],[26,9603,8205],{"className":9604},[59],[26,9606],{"className":9607,"style":55},[54],[26,9609,9611,9614,9654],{"className":9610},[38],[26,9612],{"className":9613,"style":314},[42],[26,9615,9617,9620],{"className":9616},[47],[26,9618,4018],{"className":9619},[47,179],[26,9621,9623],{"className":9622},[184],[26,9624,9626,9646],{"className":9625},[188,189],[26,9627,9629,9643],{"className":9628},[193],[26,9630,9632],{"className":9631,"style":198},[197],[26,9633,9634,9637],{"style":201},[26,9635],{"className":9636,"style":206},[205],[26,9638,9640],{"className":9639},[210,211,212,213],[26,9641,158],{"className":9642},[47,48,213],[26,9644,221],{"className":9645},[220],[26,9647,9649],{"className":9648},[193],[26,9650,9652],{"className":9651,"style":228},[197],[26,9653],{},[26,9655,765],{"className":9656},[47],[11,9658,9659,9662],{},[21,9660,9661],{},"The output gate"," decides what of the cell to expose as the hidden state used for the current\noutput and passed to the next step:",[26,9664,9666],{"className":9665},[301],[26,9667,9669],{"className":9668},[29],[26,9670,9672,9733,10002,10058],{"className":9671,"ariaHidden":34},[33],[26,9673,9675,9678,9718,9721,9724,9727,9730],{"className":9674},[38],[26,9676],{"className":9677,"style":172},[42],[26,9679,9681,9684],{"className":9680},[47],[26,9682,1168],{"className":9683},[47,179],[26,9685,9687],{"className":9686},[184],[26,9688,9690,9710],{"className":9689},[188,189],[26,9691,9693,9707],{"className":9692},[193],[26,9694,9696],{"className":9695,"style":198},[197],[26,9697,9698,9701],{"style":201},[26,9699],{"className":9700,"style":206},[205],[26,9702,9704],{"className":9703},[210,211,212,213],[26,9705,158],{"className":9706},[47,48,213],[26,9708,221],{"className":9709},[220],[26,9711,9713],{"className":9712},[193],[26,9714,9716],{"className":9715,"style":228},[197],[26,9717],{},[26,9719],{"className":9720,"style":358},[54],[26,9722],{"className":9723,"style":358},[54],[26,9725,366],{"className":9726},[365],[26,9728],{"className":9729,"style":358},[54],[26,9731],{"className":9732,"style":358},[54],[26,9734,9736,9739,9742,9745,9748,9935,9938,9941,9944,9947,9987,9990,9993,9996,9999],{"className":9735},[38],[26,9737],{"className":9738,"style":379},[42],[26,9740,8156],{"className":9741,"style":383},[47,48],[26,9743],{"className":9744,"style":388},[54],[26,9746],{"className":9747,"style":392},[54],[26,9749,9751,9754,9794,9843,9846,9849,9852,9892,9932],{"className":9750},[396],[26,9752,403],{"className":9753,"style":402},[400,401],[26,9755,9757,9760],{"className":9756},[47],[26,9758,407],{"className":9759},[47,179],[26,9761,9763],{"className":9762},[184],[26,9764,9766,9786],{"className":9765},[188,189],[26,9767,9769,9783],{"className":9768},[193],[26,9770,9772],{"className":9771,"style":1539},[197],[26,9773,9774,9777],{"style":201},[26,9775],{"className":9776,"style":206},[205],[26,9778,9780],{"className":9779},[210,211,212,213],[26,9781,1168],{"className":9782},[47,48,213],[26,9784,221],{"className":9785},[220],[26,9787,9789],{"className":9788},[193],[26,9790,9792],{"className":9791,"style":228},[197],[26,9793],{},[26,9795,9797,9800],{"className":9796},[47],[26,9798,251],{"className":9799},[47,179],[26,9801,9803],{"className":9802},[184],[26,9804,9806,9835],{"className":9805},[188,189],[26,9807,9809,9832],{"className":9808},[193],[26,9810,9812],{"className":9811,"style":264},[197],[26,9813,9814,9817],{"style":201},[26,9815],{"className":9816,"style":206},[205],[26,9818,9820],{"className":9819},[210,211,212,213],[26,9821,9823,9826,9829],{"className":9822},[47,213],[26,9824,158],{"className":9825},[47,48,213],[26,9827,60],{"className":9828},[59,213],[26,9830,74],{"className":9831},[47,213],[26,9833,221],{"className":9834},[220],[26,9836,9838],{"className":9837},[193],[26,9839,9841],{"className":9840,"style":294},[197],[26,9842],{},[26,9844],{"className":9845,"style":55},[54],[26,9847,463],{"className":9848},[59],[26,9850],{"className":9851,"style":55},[54],[26,9853,9855,9858],{"className":9854},[47],[26,9856,471],{"className":9857,"style":470},[47,179],[26,9859,9861],{"className":9860},[184],[26,9862,9864,9884],{"className":9863},[188,189],[26,9865,9867,9881],{"className":9866},[193],[26,9868,9870],{"className":9869,"style":1539},[197],[26,9871,9872,9875],{"style":650},[26,9873],{"className":9874,"style":206},[205],[26,9876,9878],{"className":9877},[210,211,212,213],[26,9879,1168],{"className":9880},[47,48,213],[26,9882,221],{"className":9883},[220],[26,9885,9887],{"className":9886},[193],[26,9888,9890],{"className":9889,"style":228},[197],[26,9891],{},[26,9893,9895,9898],{"className":9894},[47],[26,9896,180],{"className":9897},[47,179],[26,9899,9901],{"className":9900},[184],[26,9902,9904,9924],{"className":9903},[188,189],[26,9905,9907,9921],{"className":9906},[193],[26,9908,9910],{"className":9909,"style":198},[197],[26,9911,9912,9915],{"style":201},[26,9913],{"className":9914,"style":206},[205],[26,9916,9918],{"className":9917},[210,211,212,213],[26,9919,158],{"className":9920},[47,48,213],[26,9922,221],{"className":9923},[220],[26,9925,9927],{"className":9926},[193],[26,9928,9930],{"className":9929,"style":228},[197],[26,9931],{},[26,9933,516],{"className":9934,"style":402},[515,401],[26,9936],{"className":9937,"style":392},[54],[26,9939,524],{"className":9940},[523],[26,9942],{"className":9943,"style":3616},[54],[26,9945],{"className":9946,"style":392},[54],[26,9948,9950,9953],{"className":9949},[47],[26,9951,251],{"className":9952},[47,179],[26,9954,9956],{"className":9955},[184],[26,9957,9959,9979],{"className":9958},[188,189],[26,9960,9962,9976],{"className":9961},[193],[26,9963,9965],{"className":9964,"style":198},[197],[26,9966,9967,9970],{"style":201},[26,9968],{"className":9969,"style":206},[205],[26,9971,9973],{"className":9972},[210,211,212,213],[26,9974,158],{"className":9975},[47,48,213],[26,9977,221],{"className":9978},[220],[26,9980,9982],{"className":9981},[193],[26,9983,9985],{"className":9984,"style":228},[197],[26,9986],{},[26,9988],{"className":9989,"style":358},[54],[26,9991],{"className":9992,"style":358},[54],[26,9994,366],{"className":9995},[365],[26,9997],{"className":9998,"style":358},[54],[26,10000],{"className":10001,"style":358},[54],[26,10003,10005,10009,10049,10052,10055],{"className":10004},[38],[26,10006],{"className":10007,"style":10008},[42],"height:0.7333em;vertical-align:-0.15em;",[26,10010,10012,10015],{"className":10011},[47],[26,10013,1168],{"className":10014},[47,179],[26,10016,10018],{"className":10017},[184],[26,10019,10021,10041],{"className":10020},[188,189],[26,10022,10024,10038],{"className":10023},[193],[26,10025,10027],{"className":10026,"style":198},[197],[26,10028,10029,10032],{"style":201},[26,10030],{"className":10031,"style":206},[205],[26,10033,10035],{"className":10034},[210,211,212,213],[26,10036,158],{"className":10037},[47,48,213],[26,10039,221],{"className":10040},[220],[26,10042,10044],{"className":10043},[193],[26,10045,10047],{"className":10046,"style":228},[197],[26,10048],{},[26,10050],{"className":10051,"style":55},[54],[26,10053,8205],{"className":10054},[59],[26,10056],{"className":10057,"style":55},[54],[26,10059,10061,10064,10067,10070,10110,10113],{"className":10060},[38],[26,10062],{"className":10063,"style":379},[42],[26,10065,562],{"className":10066},[561],[26,10068,403],{"className":10069},[400],[26,10071,10073,10076],{"className":10072},[47],[26,10074,8089],{"className":10075},[47,179],[26,10077,10079],{"className":10078},[184],[26,10080,10082,10102],{"className":10081},[188,189],[26,10083,10085,10099],{"className":10084},[193],[26,10086,10088],{"className":10087,"style":198},[197],[26,10089,10090,10093],{"style":201},[26,10091],{"className":10092,"style":206},[205],[26,10094,10096],{"className":10095},[210,211,212,213],[26,10097,158],{"className":10098},[47,48,213],[26,10100,221],{"className":10101},[220],[26,10103,10105],{"className":10104},[193],[26,10106,10108],{"className":10107,"style":228},[197],[26,10109],{},[26,10111,516],{"className":10112},[515],[26,10114,765],{"className":10115},[47],[1319,10117],{"hash":10118},"197f74586cadd86b53416051d5e03353ce98ce76de6e6079f5c72f49d98e1d5b",[11,10120,10121,10122,10312,10313,10316,10317,10388,10389,10404,10405,10407,10408,10410],{},"The gate that matters most for the gradient is the forget gate on the cell state. The cell\nupdate ",[26,10123,10125],{"className":10124},[29],[26,10126,10128,10183,10247,10302],{"className":10127,"ariaHidden":34},[33],[26,10129,10131,10134,10174,10177,10180],{"className":10130},[38],[26,10132],{"className":10133,"style":172},[42],[26,10135,10137,10140],{"className":10136},[47],[26,10138,8089],{"className":10139},[47,179],[26,10141,10143],{"className":10142},[184],[26,10144,10146,10166],{"className":10145},[188,189],[26,10147,10149,10163],{"className":10148},[193],[26,10150,10152],{"className":10151,"style":198},[197],[26,10153,10154,10157],{"style":201},[26,10155],{"className":10156,"style":206},[205],[26,10158,10160],{"className":10159},[210,211,212,213],[26,10161,158],{"className":10162},[47,48,213],[26,10164,221],{"className":10165},[220],[26,10167,10169],{"className":10168},[193],[26,10170,10172],{"className":10171,"style":228},[197],[26,10173],{},[26,10175],{"className":10176,"style":358},[54],[26,10178,366],{"className":10179},[365],[26,10181],{"className":10182,"style":358},[54],[26,10184,10186,10189,10238,10241,10244],{"className":10185},[38],[26,10187],{"className":10188,"style":8744},[42],[26,10190,10192,10195],{"className":10191},[47],[26,10193,8089],{"className":10194},[47,179],[26,10196,10198],{"className":10197},[184],[26,10199,10201,10230],{"className":10200},[188,189],[26,10202,10204,10227],{"className":10203},[193],[26,10205,10207],{"className":10206,"style":264},[197],[26,10208,10209,10212],{"style":201},[26,10210],{"className":10211,"style":206},[205],[26,10213,10215],{"className":10214},[210,211,212,213],[26,10216,10218,10221,10224],{"className":10217},[47,213],[26,10219,158],{"className":10220},[47,48,213],[26,10222,60],{"className":10223},[59,213],[26,10225,74],{"className":10226},[47,213],[26,10228,221],{"className":10229},[220],[26,10231,10233],{"className":10232},[193],[26,10234,10236],{"className":10235,"style":294},[197],[26,10237],{},[26,10239],{"className":10240,"style":55},[54],[26,10242,8205],{"className":10243},[59],[26,10245],{"className":10246,"style":55},[54],[26,10248,10250,10253,10293,10296,10299],{"className":10249},[38],[26,10251],{"className":10252,"style":314},[42],[26,10254,10256,10259],{"className":10255},[47],[26,10257,8295],{"className":10258,"style":49},[47,179],[26,10260,10262],{"className":10261},[184],[26,10263,10265,10285],{"className":10264},[188,189],[26,10266,10268,10282],{"className":10267},[193],[26,10269,10271],{"className":10270,"style":198},[197],[26,10272,10273,10276],{"style":8310},[26,10274],{"className":10275,"style":206},[205],[26,10277,10279],{"className":10278},[210,211,212,213],[26,10280,158],{"className":10281},[47,48,213],[26,10283,221],{"className":10284},[220],[26,10286,10288],{"className":10287},[193],[26,10289,10291],{"className":10290,"style":228},[197],[26,10292],{},[26,10294],{"className":10295,"style":55},[54],[26,10297,463],{"className":10298},[59],[26,10300],{"className":10301,"style":55},[54],[26,10303,10305,10309],{"className":10304},[38],[26,10306],{"className":10307,"style":10308},[42],"height:0.123em;",[26,10310,1511],{"className":10311},[396]," is (nearly) ",[21,10314,10315],{},"additive",":\nwhen the forget gate stays open (",[26,10318,10320],{"className":10319},[29],[26,10321,10323,10379],{"className":10322,"ariaHidden":34},[33],[26,10324,10326,10329,10369,10372,10376],{"className":10325},[38],[26,10327],{"className":10328,"style":314},[42],[26,10330,10332,10335],{"className":10331},[47],[26,10333,8295],{"className":10334,"style":49},[47,179],[26,10336,10338],{"className":10337},[184],[26,10339,10341,10361],{"className":10340},[188,189],[26,10342,10344,10358],{"className":10343},[193],[26,10345,10347],{"className":10346,"style":198},[197],[26,10348,10349,10352],{"style":8310},[26,10350],{"className":10351,"style":206},[205],[26,10353,10355],{"className":10354},[210,211,212,213],[26,10356,158],{"className":10357},[47,48,213],[26,10359,221],{"className":10360},[220],[26,10362,10364],{"className":10363},[193],[26,10365,10367],{"className":10366,"style":228},[197],[26,10368],{},[26,10370],{"className":10371,"style":358},[54],[26,10373,10375],{"className":10374},[365],"≈",[26,10377],{"className":10378,"style":358},[54],[26,10380,10382,10385],{"className":10381},[38],[26,10383],{"className":10384,"style":70},[42],[26,10386,74],{"className":10387},[47],"), backpropagating through the cell\nmultiplies the gradient by roughly ",[26,10390,10392],{"className":10391},[29],[26,10393,10395],{"className":10394,"ariaHidden":34},[33],[26,10396,10398,10401],{"className":10397},[38],[26,10399],{"className":10400,"style":70},[42],[26,10402,74],{"className":10403},[47]," at each step instead of by a shrinking factor. Context can\nsit in the cell across dozens of steps with its gradient intact — the additive path is\nwhat plain RNNs lacked, and the reason the LSTM can hold ",[94,10406,8052],{}," until ",[94,10409,8044],{}," needs it.",[1440,10412,10413],{"type":1442},[11,10414,10415,10418,10419,10476,10477,10480,10481,10484,10485,10488],{},[21,10416,10417],{},"Definition (LSTM gate)."," A feedforward layer followed by a sigmoid, whose output in\n",[26,10420,10422],{"className":10421},[29],[26,10423,10425],{"className":10424,"ariaHidden":34},[33],[26,10426,10428,10432,10435,10438,10441,10444,10447],{"className":10427},[38],[26,10429],{"className":10430,"style":10431},[42],"height:1.0991em;vertical-align:-0.25em;",[26,10433,4301],{"className":10434},[400],[26,10436,2291],{"className":10437},[47],[26,10439,524],{"className":10440},[523],[26,10442],{"className":10443,"style":392},[54],[26,10445,74],{"className":10446},[47],[26,10448,10450,10453],{"className":10449},[515],[26,10451,4308],{"className":10452},[515],[26,10454,10456],{"className":10455},[184],[26,10457,10459],{"className":10458},[188],[26,10460,10462],{"className":10461},[193],[26,10463,10465],{"className":10464,"style":801},[197],[26,10466,10467,10470],{"style":824},[26,10468],{"className":10469,"style":206},[205],[26,10471,10473],{"className":10472},[210,211,212,213],[26,10474,840],{"className":10475},[47,48,213]," is multiplied element-wise into the vector it controls. The ",[21,10478,10479],{},"forget gate"," removes\ninformation from the cell, the ",[21,10482,10483],{},"input gate"," admits new information, and the ",[21,10486,10487],{},"output gate","\nexposes the cell as the hidden state. Sigmoid gates turn context management into learned,\ndifferentiable masking.",[1963,10490,10492],{"id":10491},"worked-example-the-forget-gate-holds-a-value","Worked example: the forget gate holds a value",[11,10494,10495,10496,10500,10501,10534,10535,10569],{},"Take a single cell coordinate storing the ",[10497,10498,10499],"q",{},"subject is plural"," bit as ",[26,10502,10504],{"className":10503},[29],[26,10505,10507,10525],{"className":10506,"ariaHidden":34},[33],[26,10508,10510,10513,10516,10519,10522],{"className":10509},[38],[26,10511],{"className":10512,"style":8152},[42],[26,10514,8089],{"className":10515},[47,48],[26,10517],{"className":10518,"style":358},[54],[26,10520,366],{"className":10521},[365],[26,10523],{"className":10524,"style":358},[54],[26,10526,10528,10531],{"className":10527},[38],[26,10529],{"className":10530,"style":70},[42],[26,10532,2393],{"className":10533},[47],", and\nwatch it survive intervening words. Suppose the network has learned, for this\ncoordinate, a forget gate that stays near open (",[26,10536,10538],{"className":10537},[29],[26,10539,10541,10559],{"className":10540,"ariaHidden":34},[33],[26,10542,10544,10547,10550,10553,10556],{"className":10543},[38],[26,10545],{"className":10546,"style":1461},[42],[26,10548,8295],{"className":10549,"style":8517},[47,48],[26,10551],{"className":10552,"style":358},[54],[26,10554,366],{"className":10555},[365],[26,10557],{"className":10558,"style":358},[54],[26,10560,10562,10565],{"className":10561},[38],[26,10563],{"className":10564,"style":70},[42],[26,10566,10568],{"className":10567},[47],"0.95",") while it should keep the\nvalue, and near closed only when a new subject arrives.",[11,10571,10572,10575,10576,10578,10579,10612,10613,10647,10648,10684],{},[21,10573,10574],{},"A word that should not overwrite"," (say the distractor ",[94,10577,8056],{},"). The forget gate\ncomputes ",[26,10580,10582],{"className":10581},[29],[26,10583,10585,10603],{"className":10584,"ariaHidden":34},[33],[26,10586,10588,10591,10594,10597,10600],{"className":10587},[38],[26,10589],{"className":10590,"style":1461},[42],[26,10592,8295],{"className":10593,"style":8517},[47,48],[26,10595],{"className":10596,"style":358},[54],[26,10598,366],{"className":10599},[365],[26,10601],{"className":10602,"style":358},[54],[26,10604,10606,10609],{"className":10605},[38],[26,10607],{"className":10608,"style":70},[42],[26,10610,10568],{"className":10611},[47],", and the input gate computes ",[26,10614,10616],{"className":10615},[29],[26,10617,10619,10637],{"className":10618,"ariaHidden":34},[33],[26,10620,10622,10625,10628,10631,10634],{"className":10621},[38],[26,10623],{"className":10624,"style":4014},[42],[26,10626,4018],{"className":10627},[47,48],[26,10629],{"className":10630,"style":358},[54],[26,10632,366],{"className":10633},[365],[26,10635],{"className":10636,"style":358},[54],[26,10638,10640,10643],{"className":10639},[38],[26,10641],{"className":10642,"style":70},[42],[26,10644,10646],{"className":10645},[47],"0.05"," with candidate ",[26,10649,10651],{"className":10650},[29],[26,10652,10654,10672],{"className":10653,"ariaHidden":34},[33],[26,10655,10657,10660,10663,10666,10669],{"className":10656},[38],[26,10658],{"className":10659,"style":540},[42],[26,10661,384],{"className":10662,"style":383},[47,48],[26,10664],{"className":10665,"style":358},[54],[26,10667,366],{"className":10668},[365],[26,10670],{"className":10671,"style":358},[54],[26,10673,10675,10678,10681],{"className":10674},[38],[26,10676],{"className":10677,"style":2462},[42],[26,10679,60],{"className":10680},[47],[26,10682,2214],{"className":10683},[47],"\n(a competing singular signal). The cell update is",[26,10686,10688],{"className":10687},[301],[26,10689,10691],{"className":10690},[29],[26,10692,10694,10757,10776,10838,10856,10874,10901,10931,10949,10968],{"className":10693,"ariaHidden":34},[33],[26,10695,10697,10700,10748,10751,10754],{"className":10696},[38],[26,10698],{"className":10699,"style":2334},[42],[26,10701,10703,10706],{"className":10702},[47],[26,10704,8089],{"className":10705},[47,48],[26,10707,10709],{"className":10708},[184],[26,10710,10712,10740],{"className":10711},[188,189],[26,10713,10715,10737],{"className":10714},[193],[26,10716,10718],{"className":10717,"style":1539},[197],[26,10719,10720,10723],{"style":201},[26,10721],{"className":10722,"style":206},[205],[26,10724,10726],{"className":10725},[210,211,212,213],[26,10727,10729],{"className":10728},[47,213],[26,10730,10733],{"className":10731},[47,10732,213],"text",[26,10734,10736],{"className":10735},[47,213],"new",[26,10738,221],{"className":10739},[220],[26,10741,10743],{"className":10742},[193],[26,10744,10746],{"className":10745,"style":228},[197],[26,10747],{},[26,10749],{"className":10750,"style":358},[54],[26,10752,366],{"className":10753},[365],[26,10755],{"className":10756,"style":358},[54],[26,10758,10760,10763,10766,10769,10773],{"className":10759},[38],[26,10761],{"className":10762,"style":1461},[42],[26,10764,8295],{"className":10765,"style":8517},[47,48],[26,10767],{"className":10768,"style":55},[54],[26,10770,10772],{"className":10771},[59],"⋅",[26,10774],{"className":10775,"style":55},[54],[26,10777,10779,10782,10829,10832,10835],{"className":10778},[38],[26,10780],{"className":10781,"style":10008},[42],[26,10783,10785,10788],{"className":10784},[47],[26,10786,8089],{"className":10787},[47,48],[26,10789,10791],{"className":10790},[184],[26,10792,10794,10821],{"className":10793},[188,189],[26,10795,10797,10818],{"className":10796},[193],[26,10798,10800],{"className":10799,"style":8505},[197],[26,10801,10802,10805],{"style":201},[26,10803],{"className":10804,"style":206},[205],[26,10806,10808],{"className":10807},[210,211,212,213],[26,10809,10811],{"className":10810},[47,213],[26,10812,10814],{"className":10813},[47,10732,213],[26,10815,10817],{"className":10816},[47,213],"old",[26,10819,221],{"className":10820},[220],[26,10822,10824],{"className":10823},[193],[26,10825,10827],{"className":10826,"style":228},[197],[26,10828],{},[26,10830],{"className":10831,"style":55},[54],[26,10833,463],{"className":10834},[59],[26,10836],{"className":10837,"style":55},[54],[26,10839,10841,10844,10847,10850,10853],{"className":10840},[38],[26,10842],{"className":10843,"style":4014},[42],[26,10845,4018],{"className":10846},[47,48],[26,10848],{"className":10849,"style":55},[54],[26,10851,10772],{"className":10852},[59],[26,10854],{"className":10855,"style":55},[54],[26,10857,10859,10862,10865,10868,10871],{"className":10858},[38],[26,10860],{"className":10861,"style":540},[42],[26,10863,384],{"className":10864,"style":383},[47,48],[26,10866],{"className":10867,"style":358},[54],[26,10869,366],{"className":10870},[365],[26,10872],{"className":10873,"style":358},[54],[26,10875,10877,10880,10883,10886,10889,10892,10895,10898],{"className":10876},[38],[26,10878],{"className":10879,"style":379},[42],[26,10881,10568],{"className":10882},[47],[26,10884,403],{"className":10885},[400],[26,10887,2393],{"className":10888},[47],[26,10890,516],{"className":10891},[515],[26,10893],{"className":10894,"style":55},[54],[26,10896,463],{"className":10897},[59],[26,10899],{"className":10900,"style":55},[54],[26,10902,10904,10907,10910,10913,10916,10919,10922,10925,10928],{"className":10903},[38],[26,10905],{"className":10906,"style":379},[42],[26,10908,10646],{"className":10909},[47],[26,10911,403],{"className":10912},[400],[26,10914,60],{"className":10915},[47],[26,10917,2214],{"className":10918},[47],[26,10920,516],{"className":10921},[515],[26,10923],{"className":10924,"style":358},[54],[26,10926,366],{"className":10927},[365],[26,10929],{"className":10930,"style":358},[54],[26,10932,10934,10937,10940,10943,10946],{"className":10933},[38],[26,10935],{"className":10936,"style":2462},[42],[26,10938,10568],{"className":10939},[47],[26,10941],{"className":10942,"style":55},[54],[26,10944,60],{"className":10945},[59],[26,10947],{"className":10948,"style":55},[54],[26,10950,10952,10955,10959,10962,10965],{"className":10951},[38],[26,10953],{"className":10954,"style":70},[42],[26,10956,10958],{"className":10957},[47],"0.04",[26,10960],{"className":10961,"style":358},[54],[26,10963,366],{"className":10964},[365],[26,10966],{"className":10967,"style":358},[54],[26,10969,10971,10974],{"className":10970},[38],[26,10972],{"className":10973,"style":70},[42],[26,10975,10977],{"className":10976},[47],"0.91.",[11,10979,10980,10981,10996,10997,11013,11014,11078,11079,11082,11083,11098],{},"The plural bit dropped only from ",[26,10982,10984],{"className":10983},[29],[26,10985,10987],{"className":10986,"ariaHidden":34},[33],[26,10988,10990,10993],{"className":10989},[38],[26,10991],{"className":10992,"style":70},[42],[26,10994,2393],{"className":10995},[47]," to ",[26,10998,11000],{"className":10999},[29],[26,11001,11003],{"className":11002,"ariaHidden":34},[33],[26,11004,11006,11009],{"className":11005},[38],[26,11007],{"className":11008,"style":70},[42],[26,11010,11012],{"className":11011},[47],"0.91"," — the distractor barely dented it,\nbecause the input gate admitted little of the new candidate. Repeat this across five\nintervening words with the same gates and the value decays as ",[26,11015,11017],{"className":11016},[29],[26,11018,11020,11068],{"className":11019,"ariaHidden":34},[33],[26,11021,11023,11026,11030,11059,11062,11065],{"className":11022},[38],[26,11024],{"className":11025,"style":7628},[42],[26,11027,11029],{"className":11028},[47],"0.9",[26,11031,11033,11036],{"className":11032},[47],[26,11034,7614],{"className":11035},[47],[26,11037,11039],{"className":11038},[184],[26,11040,11042],{"className":11041},[188],[26,11043,11045],{"className":11044},[193],[26,11046,11048],{"className":11047,"style":7628},[197],[26,11049,11050,11053],{"style":824},[26,11051],{"className":11052,"style":206},[205],[26,11054,11056],{"className":11055},[210,211,212,213],[26,11057,7614],{"className":11058},[47,213],[26,11060],{"className":11061,"style":358},[54],[26,11063,10375],{"className":11064},[365],[26,11066],{"className":11067,"style":358},[54],[26,11069,11071,11074],{"className":11070},[38],[26,11072],{"className":11073,"style":70},[42],[26,11075,11077],{"className":11076},[47],"0.77",",\nstill clearly positive when the verb finally needs it. Contrast a plain RNN, whose\nsingle state would be ",[94,11080,11081],{},"overwritten"," by each word: with the tanh update of the earlier\nexample, the plural signal is gone within two or three steps. The forget gate near ",[26,11084,11086],{"className":11085},[29],[26,11087,11089],{"className":11088,"ariaHidden":34},[33],[26,11090,11092,11095],{"className":11091},[38],[26,11093],{"className":11094,"style":70},[42],[26,11096,74],{"className":11097},[47],"\nis what turns the multiplicative decay into near-preservation.",[11,11100,11101,11104,11105,11295,11296,11461,11462,11532,11533,11561,11562,11589,11590,11652,11653,11718],{},[21,11102,11103],{},"The gradient view."," Because ",[26,11106,11108],{"className":11107},[29],[26,11109,11111,11166,11222,11286],{"className":11110,"ariaHidden":34},[33],[26,11112,11114,11117,11157,11160,11163],{"className":11113},[38],[26,11115],{"className":11116,"style":2334},[42],[26,11118,11120,11123],{"className":11119},[47],[26,11121,8089],{"className":11122},[47,48],[26,11124,11126],{"className":11125},[184],[26,11127,11129,11149],{"className":11128},[188,189],[26,11130,11132,11146],{"className":11131},[193],[26,11133,11135],{"className":11134,"style":198},[197],[26,11136,11137,11140],{"style":201},[26,11138],{"className":11139,"style":206},[205],[26,11141,11143],{"className":11142},[210,211,212,213],[26,11144,158],{"className":11145},[47,48,213],[26,11147,221],{"className":11148},[220],[26,11150,11152],{"className":11151},[193],[26,11153,11155],{"className":11154,"style":228},[197],[26,11156],{},[26,11158],{"className":11159,"style":358},[54],[26,11161,366],{"className":11162},[365],[26,11164],{"className":11165,"style":358},[54],[26,11167,11169,11172,11213,11216,11219],{"className":11168},[38],[26,11170],{"className":11171,"style":1461},[42],[26,11173,11175,11178],{"className":11174},[47],[26,11176,8295],{"className":11177,"style":8517},[47,48],[26,11179,11181],{"className":11180},[184],[26,11182,11184,11205],{"className":11183},[188,189],[26,11185,11187,11202],{"className":11186},[193],[26,11188,11190],{"className":11189,"style":198},[197],[26,11191,11193,11196],{"style":11192},"top:-2.55em;margin-left:-0.1076em;margin-right:0.05em;",[26,11194],{"className":11195,"style":206},[205],[26,11197,11199],{"className":11198},[210,211,212,213],[26,11200,158],{"className":11201},[47,48,213],[26,11203,221],{"className":11204},[220],[26,11206,11208],{"className":11207},[193],[26,11209,11211],{"className":11210,"style":228},[197],[26,11212],{},[26,11214],{"className":11215,"style":55},[54],[26,11217,8205],{"className":11218},[59],[26,11220],{"className":11221,"style":55},[54],[26,11223,11225,11228,11277,11280,11283],{"className":11224},[38],[26,11226],{"className":11227,"style":8744},[42],[26,11229,11231,11234],{"className":11230},[47],[26,11232,8089],{"className":11233},[47,48],[26,11235,11237],{"className":11236},[184],[26,11238,11240,11269],{"className":11239},[188,189],[26,11241,11243,11266],{"className":11242},[193],[26,11244,11246],{"className":11245,"style":264},[197],[26,11247,11248,11251],{"style":201},[26,11249],{"className":11250,"style":206},[205],[26,11252,11254],{"className":11253},[210,211,212,213],[26,11255,11257,11260,11263],{"className":11256},[47,213],[26,11258,158],{"className":11259},[47,48,213],[26,11261,60],{"className":11262},[59,213],[26,11264,74],{"className":11265},[47,213],[26,11267,221],{"className":11268},[220],[26,11270,11272],{"className":11271},[193],[26,11273,11275],{"className":11274,"style":294},[197],[26,11276],{},[26,11278],{"className":11279,"style":55},[54],[26,11281,463],{"className":11282},[59],[26,11284],{"className":11285,"style":55},[54],[26,11287,11289,11292],{"className":11288},[38],[26,11290],{"className":11291,"style":10308},[42],[26,11293,1511],{"className":11294},[396],", the backward\nderivative ",[26,11297,11299],{"className":11298},[29],[26,11300,11302,11415],{"className":11301,"ariaHidden":34},[33],[26,11303,11305,11308,11311,11351,11354,11357,11406,11409,11412],{"className":11304},[38],[26,11306],{"className":11307,"style":379},[42],[26,11309,6413],{"className":11310,"style":6412},[47],[26,11312,11314,11317],{"className":11313},[47],[26,11315,8089],{"className":11316},[47,48],[26,11318,11320],{"className":11319},[184],[26,11321,11323,11343],{"className":11322},[188,189],[26,11324,11326,11340],{"className":11325},[193],[26,11327,11329],{"className":11328,"style":198},[197],[26,11330,11331,11334],{"style":201},[26,11332],{"className":11333,"style":206},[205],[26,11335,11337],{"className":11336},[210,211,212,213],[26,11338,158],{"className":11339},[47,48,213],[26,11341,221],{"className":11342},[220],[26,11344,11346],{"className":11345},[193],[26,11347,11349],{"className":11348,"style":228},[197],[26,11350],{},[26,11352,7876],{"className":11353},[47],[26,11355,6413],{"className":11356,"style":6412},[47],[26,11358,11360,11363],{"className":11359},[47],[26,11361,8089],{"className":11362},[47,48],[26,11364,11366],{"className":11365},[184],[26,11367,11369,11398],{"className":11368},[188,189],[26,11370,11372,11395],{"className":11371},[193],[26,11373,11375],{"className":11374,"style":264},[197],[26,11376,11377,11380],{"style":201},[26,11378],{"className":11379,"style":206},[205],[26,11381,11383],{"className":11382},[210,211,212,213],[26,11384,11386,11389,11392],{"className":11385},[47,213],[26,11387,158],{"className":11388},[47,48,213],[26,11390,60],{"className":11391},[59,213],[26,11393,74],{"className":11394},[47,213],[26,11396,221],{"className":11397},[220],[26,11399,11401],{"className":11400},[193],[26,11402,11404],{"className":11403,"style":294},[197],[26,11405],{},[26,11407],{"className":11408,"style":358},[54],[26,11410,366],{"className":11411},[365],[26,11413],{"className":11414,"style":358},[54],[26,11416,11418,11421],{"className":11417},[38],[26,11419],{"className":11420,"style":1461},[42],[26,11422,11424,11427],{"className":11423},[47],[26,11425,8295],{"className":11426,"style":8517},[47,48],[26,11428,11430],{"className":11429},[184],[26,11431,11433,11453],{"className":11432},[188,189],[26,11434,11436,11450],{"className":11435},[193],[26,11437,11439],{"className":11438,"style":198},[197],[26,11440,11441,11444],{"style":11192},[26,11442],{"className":11443,"style":206},[205],[26,11445,11447],{"className":11446},[210,211,212,213],[26,11448,158],{"className":11449},[47,48,213],[26,11451,221],{"className":11452},[220],[26,11454,11456],{"className":11455},[193],[26,11457,11459],{"className":11458,"style":228},[197],[26,11460],{},". When ",[26,11463,11465],{"className":11464},[29],[26,11466,11468,11523],{"className":11467,"ariaHidden":34},[33],[26,11469,11471,11474,11514,11517,11520],{"className":11470},[38],[26,11472],{"className":11473,"style":1461},[42],[26,11475,11477,11480],{"className":11476},[47],[26,11478,8295],{"className":11479,"style":8517},[47,48],[26,11481,11483],{"className":11482},[184],[26,11484,11486,11506],{"className":11485},[188,189],[26,11487,11489,11503],{"className":11488},[193],[26,11490,11492],{"className":11491,"style":198},[197],[26,11493,11494,11497],{"style":11192},[26,11495],{"className":11496,"style":206},[205],[26,11498,11500],{"className":11499},[210,211,212,213],[26,11501,158],{"className":11502},[47,48,213],[26,11504,221],{"className":11505},[220],[26,11507,11509],{"className":11508},[193],[26,11510,11512],{"className":11511,"style":228},[197],[26,11513],{},[26,11515],{"className":11516,"style":358},[54],[26,11518,10375],{"className":11519},[365],[26,11521],{"className":11522,"style":358},[54],[26,11524,11526,11529],{"className":11525},[38],[26,11527],{"className":11528,"style":70},[42],[26,11530,74],{"className":11531},[47]," the gradient\ncrossing the cell is multiplied by ",[26,11534,11536],{"className":11535},[29],[26,11537,11539,11552],{"className":11538,"ariaHidden":34},[33],[26,11540,11542,11546,11549],{"className":11541},[38],[26,11543],{"className":11544,"style":11545},[42],"height:0.4831em;",[26,11547,10375],{"className":11548},[365],[26,11550],{"className":11551,"style":358},[54],[26,11553,11555,11558],{"className":11554},[38],[26,11556],{"className":11557,"style":70},[42],[26,11559,74],{"className":11560},[47]," per step instead of by the ",[26,11563,11565],{"className":11564},[29],[26,11566,11568,11580],{"className":11567,"ariaHidden":34},[33],[26,11569,11571,11574,11577],{"className":11570},[38],[26,11572],{"className":11573,"style":11545},[42],[26,11575,10375],{"className":11576},[365],[26,11578],{"className":11579,"style":358},[54],[26,11581,11583,11586],{"className":11582},[38],[26,11584],{"className":11585,"style":70},[42],[26,11587,7565],{"className":11588},[47],"\nof the plain-RNN example — so ",[26,11591,11593],{"className":11592},[29],[26,11594,11596,11643],{"className":11595,"ariaHidden":34},[33],[26,11597,11599,11602,11634,11637,11640],{"className":11598},[38],[26,11600],{"className":11601,"style":7628},[42],[26,11603,11605,11608],{"className":11604},[47],[26,11606,74],{"className":11607},[47],[26,11609,11611],{"className":11610},[184],[26,11612,11614],{"className":11613},[188],[26,11615,11617],{"className":11616},[193],[26,11618,11620],{"className":11619,"style":7628},[197],[26,11621,11622,11625],{"style":824},[26,11623],{"className":11624,"style":206},[205],[26,11626,11628],{"className":11627},[210,211,212,213],[26,11629,11631],{"className":11630},[47,213],[26,11632,7781],{"className":11633},[47,213],[26,11635],{"className":11636,"style":358},[54],[26,11638,366],{"className":11639},[365],[26,11641],{"className":11642,"style":358},[54],[26,11644,11646,11649],{"className":11645},[38],[26,11647],{"className":11648,"style":70},[42],[26,11650,74],{"className":11651},[47]," where ",[26,11654,11656],{"className":11655},[29],[26,11657,11659,11709],{"className":11658,"ariaHidden":34},[33],[26,11660,11662,11665,11668,11700,11703,11706],{"className":11661},[38],[26,11663],{"className":11664,"style":7628},[42],[26,11666,7632],{"className":11667},[47],[26,11669,11671,11674],{"className":11670},[47],[26,11672,7639],{"className":11673},[47],[26,11675,11677],{"className":11676},[184],[26,11678,11680],{"className":11679},[188],[26,11681,11683],{"className":11682},[193],[26,11684,11686],{"className":11685,"style":7628},[197],[26,11687,11688,11691],{"style":824},[26,11689],{"className":11690,"style":206},[205],[26,11692,11694],{"className":11693},[210,211,212,213],[26,11695,11697],{"className":11696},[47,213],[26,11698,7781],{"className":11699},[47,213],[26,11701],{"className":11702,"style":358},[54],[26,11704,366],{"className":11705},[365],[26,11707],{"className":11708,"style":358},[54],[26,11710,11712,11715],{"className":11711},[38],[26,11713],{"className":11714,"style":70},[42],[26,11716,7847],{"className":11717},[47],". The cell is a\ngradient-preserving path, and the forget gate controls it.",[11,11720,11721,11722,11726,11727,765],{},"The full derivation of the gates, their gradients, and the additive cell-state path is in the\ndeep-learning course's ",[15,11723,11725],{"href":11724},"\u002Fdeep-learning\u002Farchitectures\u002Flstm-and-gru","LSTM & GRU"," lesson; the RNN\nrecurrence and BPTT are treated in\n",[15,11728,11730],{"href":11729},"\u002Fdeep-learning\u002Farchitectures\u002Frecurrent-networks","recurrent networks",[1963,11732,11734],{"id":11733},"the-gru","The GRU",[11,11736,11737,11738,11741,11742,11745,11746,11749],{},"The LSTM's three gates and separate cell state are effective but heavy — four feedforward layers\nper unit. The ",[21,11739,11740],{},"gated recurrent unit"," (GRU) is a lighter design that keeps the same\ngradient-preserving idea with fewer parts: it drops the separate cell state and uses just two\ngates, a ",[94,11743,11744],{},"reset"," gate that controls how much past state enters the candidate and an ",[94,11747,11748],{},"update"," gate\nthat interpolates between the old state and the candidate. It trains faster and often matches the\nLSTM's accuracy, so it is a common default when the LSTM's full capacity is not needed.",[127,11751,11753],{"id":11752},"the-lstm-lineage-and-the-rise-of-attention","The LSTM lineage and the rise of attention",[11,11755,11756],{},"The recurrent architectures of this lesson have a precise public history, and the last\nstep in it is the one that made them obsolete for translation.",[11,11758,11759,11761,11762,11769,11770,11773,11774,11789],{},[21,11760,8061],{}," (Hochreiter and Schmidhuber, 1997).",[117,11763,11764],{},[15,11765,7639],{"href":11766,"ariaDescribedBy":11767,"dataFootnoteRef":6,"id":11768},"#user-content-fn-hochreiter",[123],"user-content-fnref-hochreiter"," The long short-term\nmemory unit was introduced two decades before it became standard, specifically to solve\nthe vanishing-gradient problem this lesson diagnoses. Their analysis showed that a plain\nrecurrent unit's error signal decays or explodes exponentially in the time lag, and\ntheir fix — the ",[10497,11771,11772],{},"constant error carousel,"," an internal cell whose self-connection has\nweight ",[26,11775,11777],{"className":11776},[29],[26,11778,11780],{"className":11779,"ariaHidden":34},[33],[26,11781,11783,11786],{"className":11782},[38],[26,11784],{"className":11785,"style":70},[42],[26,11787,74],{"className":11788},[47]," so the gradient neither grows nor shrinks — coincides with the additive cell\npath above. The forget gate was added later (Gers, Schmidhuber, and Cummins, 2000) and is now\npart of the standard cell.",[11,11791,11792,11795,11796,11804,11805,11808],{},[21,11793,11794],{},"Sequence-to-sequence and the GRU"," (Sutskever, Vinyals, and Le, 2014; Cho et al.,\n2014).",[117,11797,11798],{},[15,11799,11803],{"href":11800,"ariaDescribedBy":11801,"dataFootnoteRef":6,"id":11802},"#user-content-fn-seq2seq",[123],"user-content-fnref-seq2seq","7"," Two 2014 papers turned stacked LSTMs into machine-translation systems.\nSutskever and colleagues trained a deep LSTM encoder-decoder end to end on English-French\nand reached a BLEU score competitive with phrase-based statistical systems; one trick\ncarried much of the gain — ",[94,11806,11807],{},"reversing"," the source sentence so early source words sat near the decoder.\nCho and colleagues introduced the GRU in the same encoder-decoder setting. Both hit the\nsingle-context-vector bottleneck this lesson closes on.",[11,11810,11811,11814,11815,11823,11824,11827,11828,11831,11832,11836],{},[21,11812,11813],{},"Attention"," (Bahdanau, Cho, and Bengio, 2015).",[117,11816,11817],{},[15,11818,11822],{"href":11819,"ariaDescribedBy":11820,"dataFootnoteRef":6,"id":11821},"#user-content-fn-bahdanau",[123],"user-content-fnref-bahdanau","8"," The fix for the bottleneck\nwas to let the decoder read a ",[94,11825,11826],{},"weighted combination of all"," encoder hidden states at each\nstep, with the weights computed from how well each source position matches the current\ndecoder state. Bahdanau and colleagues showed this ",[10497,11829,11830],{},"align and translate"," mechanism\nremoved the sharp quality drop on long sentences that the fixed-vector encoder-decoder\nsuffered. Attention was first an add-on to an RNN; the ",[15,11833,11835],{"href":11834},"\u002Fnatural-language-processing\u002Ftransformers\u002Ftransformers-and-attention","transformer","\n(Vaswani et al., 2017) then kept the attention and discarded the recurrence entirely,\nregaining the parallelism across time that the sequential RNN gave up — the exact\nweakness noted after the forward-inference algorithm above.",[127,11838,11840],{"id":11839},"stacked-and-bidirectional-rnns","Stacked and bidirectional RNNs",[11,11842,11843],{},"The RNN cell is a module — vectors in, vectors out — so it composes in two useful ways.",[11,11845,11846,11847,11850],{},"A ",[21,11848,11849],{},"stacked RNN"," feeds the entire output sequence of one RNN as the input sequence of another,\npiling several layers deep. Lower layers induce local, concrete features; higher layers build\nmore abstract representations on top, the same layered-abstraction argument that motivates depth\nelsewhere. Stacked RNNs generally outperform single-layer ones, though training cost rises with\ndepth.",[11,11852,11846,11853,11856,11857,11872,11873,11888,11889,11904],{},[21,11854,11855],{},"bidirectional RNN"," addresses a different limitation: a left-to-right RNN's state at step ",[26,11858,11860],{"className":11859},[29],[26,11861,11863],{"className":11862,"ariaHidden":34},[33],[26,11864,11866,11869],{"className":11865},[38],[26,11867],{"className":11868,"style":154},[42],[26,11870,158],{"className":11871},[47,48],"\nknows only the words up to ",[26,11874,11876],{"className":11875},[29],[26,11877,11879],{"className":11878,"ariaHidden":34},[33],[26,11880,11882,11885],{"className":11881},[38],[26,11883],{"className":11884,"style":154},[42],[26,11886,158],{"className":11887},[47,48],", nothing to the right. For tasks where the whole input is\navailable — labeling a token given its full sentence, classifying a complete document — the words\nafter ",[26,11890,11892],{"className":11891},[29],[26,11893,11895],{"className":11894,"ariaHidden":34},[33],[26,11896,11898,11901],{"className":11897},[38],[26,11899],{"className":11900,"style":154},[42],[26,11902,158],{"className":11903},[47,48]," are informative too. A bidirectional RNN runs two independent RNNs, one forward and one\nbackward over the reversed sequence, and concatenates their states at each position:",[26,11906,11908],{"className":11907},[301],[26,11909,11911],{"className":11910},[29],[26,11912,11914,11988,12227,12445,12519],{"className":11913,"ariaHidden":34},[33],[26,11915,11917,11921,11979,11982,11985],{"className":11916},[38],[26,11918],{"className":11919,"style":11920},[42],"height:1.2128em;vertical-align:-0.2458em;",[26,11922,11924,11927],{"className":11923},[47],[26,11925,251],{"className":11926},[47,179],[26,11928,11930],{"className":11929},[184],[26,11931,11933,11970],{"className":11932},[188,189],[26,11934,11936,11967],{"className":11935},[193],[26,11937,11940,11952],{"className":11938,"style":11939},[197],"height:0.967em;",[26,11941,11943,11946],{"style":11942},"top:-2.4542em;margin-left:0em;margin-right:0.05em;",[26,11944],{"className":11945,"style":206},[205],[26,11947,11949],{"className":11948},[210,211,212,213],[26,11950,158],{"className":11951},[47,48,213],[26,11953,11955,11958],{"style":11954},"top:-3.1809em;margin-right:0.05em;",[26,11956],{"className":11957,"style":206},[205],[26,11959,11961],{"className":11960},[210,211,212,213],[26,11962,11964],{"className":11963},[47,213],[26,11965,8295],{"className":11966,"style":8517},[47,48,213],[26,11968,221],{"className":11969},[220],[26,11971,11973],{"className":11972},[193],[26,11974,11977],{"className":11975,"style":11976},[197],"height:0.2458em;",[26,11978],{},[26,11980],{"className":11981,"style":358},[54],[26,11983,366],{"className":11984},[365],[26,11986],{"className":11987,"style":358},[54],[26,11989,11991,11995,12047,12050,12090,12093,12096,12099,12102,12105,12108,12148,12151,12154,12157,12160,12218,12221,12224],{"className":11990},[38],[26,11992],{"className":11993,"style":11994},[42],"height:1.1491em;vertical-align:-0.25em;",[26,11996,11998,12005],{"className":11997},[47],[26,11999,12001],{"className":12000},[47],[26,12002,12004],{"className":12003},[47,698],"RNN",[26,12006,12008],{"className":12007},[184],[26,12009,12011,12039],{"className":12010},[188,189],[26,12012,12014,12036],{"className":12013},[193],[26,12015,12017],{"className":12016,"style":8505},[197],[26,12018,12020,12023],{"style":12019},"top:-2.55em;margin-right:0.05em;",[26,12021],{"className":12022,"style":206},[205],[26,12024,12026],{"className":12025},[210,211,212,213],[26,12027,12029],{"className":12028},[47,213],[26,12030,12032],{"className":12031},[47,10732,213],[26,12033,12035],{"className":12034},[47,213],"forward",[26,12037,221],{"className":12038},[220],[26,12040,12042],{"className":12041},[193],[26,12043,12045],{"className":12044,"style":228},[197],[26,12046],{},[26,12048,403],{"className":12049},[400],[26,12051,12053,12056],{"className":12052},[47],[26,12054,180],{"className":12055},[47,179],[26,12057,12059],{"className":12058},[184],[26,12060,12062,12082],{"className":12061},[188,189],[26,12063,12065,12079],{"className":12064},[193],[26,12066,12068],{"className":12067,"style":264},[197],[26,12069,12070,12073],{"style":201},[26,12071],{"className":12072,"style":206},[205],[26,12074,12076],{"className":12075},[210,211,212,213],[26,12077,74],{"className":12078},[47,213],[26,12080,221],{"className":12081},[220],[26,12083,12085],{"className":12084},[193],[26,12086,12088],{"className":12087,"style":228},[197],[26,12089],{},[26,12091,524],{"className":12092},[523],[26,12094],{"className":12095,"style":392},[54],[26,12097,1511],{"className":12098},[396],[26,12100],{"className":12101,"style":392},[54],[26,12103,524],{"className":12104},[523],[26,12106],{"className":12107,"style":392},[54],[26,12109,12111,12114],{"className":12110},[47],[26,12112,180],{"className":12113},[47,179],[26,12115,12117],{"className":12116},[184],[26,12118,12120,12140],{"className":12119},[188,189],[26,12121,12123,12137],{"className":12122},[193],[26,12124,12126],{"className":12125,"style":198},[197],[26,12127,12128,12131],{"style":201},[26,12129],{"className":12130,"style":206},[205],[26,12132,12134],{"className":12133},[210,211,212,213],[26,12135,158],{"className":12136},[47,48,213],[26,12138,221],{"className":12139},[220],[26,12141,12143],{"className":12142},[193],[26,12144,12146],{"className":12145,"style":228},[197],[26,12147],{},[26,12149,516],{"className":12150},[515],[26,12152,524],{"className":12153},[523],[26,12155],{"className":12156,"style":3616},[54],[26,12158],{"className":12159,"style":392},[54],[26,12161,12163,12166],{"className":12162},[47],[26,12164,251],{"className":12165},[47,179],[26,12167,12169],{"className":12168},[184],[26,12170,12172,12209],{"className":12171},[188,189],[26,12173,12175,12206],{"className":12174},[193],[26,12176,12179,12191],{"className":12177,"style":12178},[197],"height:0.8991em;",[26,12180,12182,12185],{"style":12181},"top:-2.453em;margin-left:0em;margin-right:0.05em;",[26,12183],{"className":12184,"style":206},[205],[26,12186,12188],{"className":12187},[210,211,212,213],[26,12189,158],{"className":12190},[47,48,213],[26,12192,12193,12196],{"style":7197},[26,12194],{"className":12195,"style":206},[205],[26,12197,12199],{"className":12198},[210,211,212,213],[26,12200,12202],{"className":12201},[47,213],[26,12203,12205],{"className":12204},[47,48,213],"b",[26,12207,221],{"className":12208},[220],[26,12210,12212],{"className":12211},[193],[26,12213,12216],{"className":12214,"style":12215},[197],"height:0.247em;",[26,12217],{},[26,12219],{"className":12220,"style":358},[54],[26,12222,366],{"className":12223},[365],[26,12225],{"className":12226,"style":358},[54],[26,12228,12230,12233,12283,12286,12326,12329,12332,12335,12338,12341,12344,12384,12387,12390,12393,12396,12436,12439,12442],{"className":12229},[38],[26,12231],{"className":12232,"style":379},[42],[26,12234,12236,12242],{"className":12235},[47],[26,12237,12239],{"className":12238},[47],[26,12240,12004],{"className":12241},[47,698],[26,12243,12245],{"className":12244},[184],[26,12246,12248,12275],{"className":12247},[188,189],[26,12249,12251,12272],{"className":12250},[193],[26,12252,12254],{"className":12253,"style":8505},[197],[26,12255,12256,12259],{"style":12019},[26,12257],{"className":12258,"style":206},[205],[26,12260,12262],{"className":12261},[210,211,212,213],[26,12263,12265],{"className":12264},[47,213],[26,12266,12268],{"className":12267},[47,10732,213],[26,12269,12271],{"className":12270},[47,213],"backward",[26,12273,221],{"className":12274},[220],[26,12276,12278],{"className":12277},[193],[26,12279,12281],{"className":12280,"style":228},[197],[26,12282],{},[26,12284,403],{"className":12285},[400],[26,12287,12289,12292],{"className":12288},[47],[26,12290,180],{"className":12291},[47,179],[26,12293,12295],{"className":12294},[184],[26,12296,12298,12318],{"className":12297},[188,189],[26,12299,12301,12315],{"className":12300},[193],[26,12302,12304],{"className":12303,"style":198},[197],[26,12305,12306,12309],{"style":201},[26,12307],{"className":12308,"style":206},[205],[26,12310,12312],{"className":12311},[210,211,212,213],[26,12313,158],{"className":12314},[47,48,213],[26,12316,221],{"className":12317},[220],[26,12319,12321],{"className":12320},[193],[26,12322,12324],{"className":12323,"style":228},[197],[26,12325],{},[26,12327,524],{"className":12328},[523],[26,12330],{"className":12331,"style":392},[54],[26,12333,1511],{"className":12334},[396],[26,12336],{"className":12337,"style":392},[54],[26,12339,524],{"className":12340},[523],[26,12342],{"className":12343,"style":392},[54],[26,12345,12347,12350],{"className":12346},[47],[26,12348,180],{"className":12349},[47,179],[26,12351,12353],{"className":12352},[184],[26,12354,12356,12376],{"className":12355},[188,189],[26,12357,12359,12373],{"className":12358},[193],[26,12360,12362],{"className":12361,"style":1539},[197],[26,12363,12364,12367],{"style":201},[26,12365],{"className":12366,"style":206},[205],[26,12368,12370],{"className":12369},[210,211,212,213],[26,12371,1551],{"className":12372},[47,48,213],[26,12374,221],{"className":12375},[220],[26,12377,12379],{"className":12378},[193],[26,12380,12382],{"className":12381,"style":228},[197],[26,12383],{},[26,12385,516],{"className":12386},[515],[26,12388,524],{"className":12389},[523],[26,12391],{"className":12392,"style":3616},[54],[26,12394],{"className":12395,"style":392},[54],[26,12397,12399,12402],{"className":12398},[47],[26,12400,251],{"className":12401},[47,179],[26,12403,12405],{"className":12404},[184],[26,12406,12408,12428],{"className":12407},[188,189],[26,12409,12411,12425],{"className":12410},[193],[26,12412,12414],{"className":12413,"style":198},[197],[26,12415,12416,12419],{"style":201},[26,12417],{"className":12418,"style":206},[205],[26,12420,12422],{"className":12421},[210,211,212,213],[26,12423,158],{"className":12424},[47,48,213],[26,12426,221],{"className":12427},[220],[26,12429,12431],{"className":12430},[193],[26,12432,12434],{"className":12433,"style":228},[197],[26,12435],{},[26,12437],{"className":12438,"style":358},[54],[26,12440,366],{"className":12441},[365],[26,12443],{"className":12444,"style":358},[54],[26,12446,12448,12452,12455,12509,12512,12516],{"className":12447},[38],[26,12449],{"className":12450,"style":12451},[42],"height:1.217em;vertical-align:-0.25em;",[26,12453,4301],{"className":12454},[400],[26,12456,12458,12461],{"className":12457},[47],[26,12459,251],{"className":12460},[47,179],[26,12462,12464],{"className":12463},[184],[26,12465,12467,12501],{"className":12466},[188,189],[26,12468,12470,12498],{"className":12469},[193],[26,12471,12473,12484],{"className":12472,"style":11939},[197],[26,12474,12475,12478],{"style":11942},[26,12476],{"className":12477,"style":206},[205],[26,12479,12481],{"className":12480},[210,211,212,213],[26,12482,158],{"className":12483},[47,48,213],[26,12485,12486,12489],{"style":11954},[26,12487],{"className":12488,"style":206},[205],[26,12490,12492],{"className":12491},[210,211,212,213],[26,12493,12495],{"className":12494},[47,213],[26,12496,8295],{"className":12497,"style":8517},[47,48,213],[26,12499,221],{"className":12500},[220],[26,12502,12504],{"className":12503},[193],[26,12505,12507],{"className":12506,"style":11976},[197],[26,12508],{},[26,12510],{"className":12511,"style":55},[54],[26,12513,12515],{"className":12514},[59],"⊕",[26,12517],{"className":12518,"style":55},[54],[26,12520,12522,12525,12579,12582],{"className":12521},[38],[26,12523],{"className":12524,"style":11994},[42],[26,12526,12528,12531],{"className":12527},[47],[26,12529,251],{"className":12530},[47,179],[26,12532,12534],{"className":12533},[184],[26,12535,12537,12571],{"className":12536},[188,189],[26,12538,12540,12568],{"className":12539},[193],[26,12541,12543,12554],{"className":12542,"style":12178},[197],[26,12544,12545,12548],{"style":12181},[26,12546],{"className":12547,"style":206},[205],[26,12549,12551],{"className":12550},[210,211,212,213],[26,12552,158],{"className":12553},[47,48,213],[26,12555,12556,12559],{"style":7197},[26,12557],{"className":12558,"style":206},[205],[26,12560,12562],{"className":12561},[210,211,212,213],[26,12563,12565],{"className":12564},[47,213],[26,12566,12205],{"className":12567},[47,48,213],[26,12569,221],{"className":12570},[220],[26,12572,12574],{"className":12573},[193],[26,12575,12577],{"className":12576,"style":12215},[197],[26,12578],{},[26,12580,4308],{"className":12581},[515],[26,12583,765],{"className":12584},[47],[11,12586,12587,12588,12640,12641,12656],{},"The combined ",[26,12589,12591],{"className":12590},[29],[26,12592,12594],{"className":12593,"ariaHidden":34},[33],[26,12595,12597,12600],{"className":12596},[38],[26,12598],{"className":12599,"style":314},[42],[26,12601,12603,12606],{"className":12602},[47],[26,12604,251],{"className":12605},[47,179],[26,12607,12609],{"className":12608},[184],[26,12610,12612,12632],{"className":12611},[188,189],[26,12613,12615,12629],{"className":12614},[193],[26,12616,12618],{"className":12617,"style":198},[197],[26,12619,12620,12623],{"style":201},[26,12621],{"className":12622,"style":206},[205],[26,12624,12626],{"className":12625},[210,211,212,213],[26,12627,158],{"className":12628},[47,48,213],[26,12630,221],{"className":12631},[220],[26,12633,12635],{"className":12634},[193],[26,12636,12638],{"className":12637,"style":228},[197],[26,12639],{}," sees both the left and right context of position ",[26,12642,12644],{"className":12643},[29],[26,12645,12647],{"className":12646,"ariaHidden":34},[33],[26,12648,12650,12653],{"className":12649},[38],[26,12651],{"className":12652,"style":154},[42],[26,12654,158],{"className":12655},[47,48]," — the\nview a local labeling decision wants.",[1319,12658],{"hash":12659},"0288c11dbf62fd8adec35ef80e7ace60b0d19a96d211e07d83b7e9a5bceaed1b",[127,12661,12663],{"id":12662},"the-encoder-decoder-model","The encoder-decoder model",[11,12665,12666,12667,12670,12671,137,12674,12677,12678],{},"The tasks so far map an input sequence to an output of the ",[94,12668,12669],{},"same length"," — one tag per word, one\nlabel per document. Machine translation, summarization, and question answering are different: the\noutput is a new sequence whose length and word order need not match the input at all. The\narchitecture built for this is the ",[21,12672,12673],{},"encoder-decoder",[21,12675,12676],{},"sequence-to-sequence",", model.",[117,12679,12680],{},[15,12681,12685],{"href":12682,"ariaDescribedBy":12683,"dataFootnoteRef":6,"id":12684},"#user-content-fn-jm-seq2seq",[123],"user-content-fnref-jm-seq2seq","9",[11,12687,12688,12689,12692,12693,12696,12697,12700],{},"It is two RNNs joined at a single vector. The ",[21,12690,12691],{},"encoder"," reads the entire source sequence and\ncompresses it into a fixed-length ",[21,12694,12695],{},"context vector"," — typically its final hidden state. The\n",[21,12698,12699],{},"decoder"," is an autoregressive RNN language model whose initial state is that context vector; it\ngenerates the target sequence one token at a time, each step conditioned on the context and the\nwords produced so far, until it emits an end symbol.",[1319,12702],{"hash":12703},"c75307818047ae227237d25d50490fe7716c22943f06205cc2ef9458bffd42d6",[11,12705,12706,12707,12726,12727,12730],{},"This design trains end-to-end on paired sequences and produced the first strong neural machine\ntranslation systems. But it has a structural flaw visible in the figure: ",[21,12708,12709,12710],{},"everything the decoder\nlearns about the source must squeeze through the single context vector ",[26,12711,12713],{"className":12712},[29],[26,12714,12716],{"className":12715,"ariaHidden":34},[33],[26,12717,12719,12723],{"className":12718},[38],[26,12720],{"className":12721,"style":12722},[42],"height:0.4444em;",[26,12724,8089],{"className":12725},[47,179],". For a short\nsentence that is fine; for a long paragraph, one fixed-length vector cannot hold every detail, and\ntranslation quality falls as the source grows. The information ",[21,12728,12729],{},"bottleneck"," is inherent to\ncompressing an arbitrarily long input into one vector.",[11,12732,12733,12734,12737,12738,12741,12742,12744],{},"The fix is to let the decoder look back at ",[94,12735,12736],{},"all"," of the encoder's hidden states, not just the final\none, weighting them by relevance at each decoding step — the mechanism called ",[21,12739,12740],{},"attention",". Attention\nremoves the bottleneck, and once a model is built entirely out of attention with the recurrence\nremoved, it becomes the ",[15,12743,11835],{"href":11834},",\nthe subject of the next module.",[12746,12747,12750,12755],"section",{"className":12748,"dataFootnotes":6},[12749],"footnotes",[127,12751,12754],{"className":12752,"id":123},[12753],"sr-only","Footnotes",[12756,12757,12758,12963,12974,12985,12996,13007,13029,13051,13073],"ol",{},[12759,12760,12762,1075,12765,12768,12769,12955,12956],"li",{"id":12761},"user-content-fn-jm-rnn",[21,12763,12764],{},"Jurafsky & Martin",[94,12766,12767],{},"Speech and Language Processing"," (3rd ed.), §9.2 — Recurrent Neural Networks: the Elman\u002Fsimple recurrent network, the recurrence ",[26,12770,12772],{"className":12771},[29],[26,12773,12775,12830,12903],{"className":12774,"ariaHidden":34},[33],[26,12776,12778,12781,12821,12824,12827],{"className":12777},[38],[26,12779],{"className":12780,"style":314},[42],[26,12782,12784,12787],{"className":12783},[47],[26,12785,251],{"className":12786},[47,179],[26,12788,12790],{"className":12789},[184],[26,12791,12793,12813],{"className":12792},[188,189],[26,12794,12796,12810],{"className":12795},[193],[26,12797,12799],{"className":12798,"style":198},[197],[26,12800,12801,12804],{"style":201},[26,12802],{"className":12803,"style":206},[205],[26,12805,12807],{"className":12806},[210,211,212,213],[26,12808,158],{"className":12809},[47,48,213],[26,12811,221],{"className":12812},[220],[26,12814,12816],{"className":12815},[193],[26,12817,12819],{"className":12818,"style":228},[197],[26,12820],{},[26,12822],{"className":12823,"style":358},[54],[26,12825,366],{"className":12826},[365],[26,12828],{"className":12829,"style":358},[54],[26,12831,12833,12836,12839,12842,12845,12894,12897,12900],{"className":12832},[38],[26,12834],{"className":12835,"style":379},[42],[26,12837,384],{"className":12838,"style":383},[47,48],[26,12840,403],{"className":12841},[400],[26,12843,407],{"className":12844},[47,179],[26,12846,12848,12851],{"className":12847},[47],[26,12849,251],{"className":12850},[47,179],[26,12852,12854],{"className":12853},[184],[26,12855,12857,12886],{"className":12856},[188,189],[26,12858,12860,12883],{"className":12859},[193],[26,12861,12863],{"className":12862,"style":264},[197],[26,12864,12865,12868],{"style":201},[26,12866],{"className":12867,"style":206},[205],[26,12869,12871],{"className":12870},[210,211,212,213],[26,12872,12874,12877,12880],{"className":12873},[47,213],[26,12875,158],{"className":12876},[47,48,213],[26,12878,60],{"className":12879},[59,213],[26,12881,74],{"className":12882},[47,213],[26,12884,221],{"className":12885},[220],[26,12887,12889],{"className":12888},[193],[26,12890,12892],{"className":12891,"style":294},[197],[26,12893],{},[26,12895],{"className":12896,"style":55},[54],[26,12898,463],{"className":12899},[59],[26,12901],{"className":12902,"style":55},[54],[26,12904,12906,12909,12912,12952],{"className":12905},[38],[26,12907],{"className":12908,"style":379},[42],[26,12910,471],{"className":12911,"style":470},[47,179],[26,12913,12915,12918],{"className":12914},[47],[26,12916,180],{"className":12917},[47,179],[26,12919,12921],{"className":12920},[184],[26,12922,12924,12944],{"className":12923},[188,189],[26,12925,12927,12941],{"className":12926},[193],[26,12928,12930],{"className":12929,"style":198},[197],[26,12931,12932,12935],{"style":201},[26,12933],{"className":12934,"style":206},[205],[26,12936,12938],{"className":12937},[210,211,212,213],[26,12939,158],{"className":12940},[47,48,213],[26,12942,221],{"className":12943},[220],[26,12945,12947],{"className":12946},[193],[26,12948,12950],{"className":12949,"style":228},[197],[26,12951],{},[26,12953,516],{"className":12954},[515],", weight sharing across time, and unrolling the cycle into a feedforward graph. ",[15,12957,12962],{"href":12958,"ariaLabel":12959,"className":12960,"dataFootnoteBackref":6},"#user-content-fnref-jm-rnn","Back to reference 1",[12961],"data-footnote-backref","↩",[12759,12964,12966,12968,12969],{"id":12965},"user-content-fn-jm-lm",[21,12967,12764],{},", §9.3 — RNNs as Language Models: next-word prediction with no fixed context window, the embedding-hidden-softmax forward pass, sequence probability as a product of per-step probabilities, and the cross-entropy training objective. ",[15,12970,12962],{"href":12971,"ariaLabel":12972,"className":12973,"dataFootnoteBackref":6},"#user-content-fnref-jm-lm","Back to reference 2",[12961],[12759,12975,12977,12979,12980],{"id":12976},"user-content-fn-jm-bptt",[21,12978,12764],{},", §9.2.2 — Training: the two-pass backpropagation-through-time algorithm, saving hidden states forward and accumulating gradients backward through time, with the three weight matrices shared across steps. ",[15,12981,12962],{"href":12982,"ariaLabel":12983,"className":12984,"dataFootnoteBackref":6},"#user-content-fnref-jm-bptt","Back to reference 3",[12961],[12759,12986,12988,12990,12991],{"id":12987},"user-content-fn-jm-vanish",[21,12989,12764],{},", §9.6 — The LSTM (motivation): the difficulty of carrying distant information, the repeated multiplications during the backward pass that drive gradients to zero, and the vanishing-gradient problem. ",[15,12992,12962],{"href":12993,"ariaLabel":12994,"className":12995,"dataFootnoteBackref":6},"#user-content-fnref-jm-vanish","Back to reference 4",[12961],[12759,12997,12999,13001,13002],{"id":12998},"user-content-fn-jm-lstm",[21,13000,12764],{},", §9.6 — The LSTM: the explicit cell\u002Fcontext layer, the forget, input (add), and output gates as sigmoid-masked feedforward layers, and the additive cell update that lets gradients flow across many steps. ",[15,13003,12962],{"href":13004,"ariaLabel":13005,"className":13006,"dataFootnoteBackref":6},"#user-content-fnref-jm-lstm","Back to reference 5",[12961],[12759,13008,13010,1075,13013,13016,13017,1075,13020,13023,13024],{"id":13009},"user-content-fn-hochreiter",[21,13011,13012],{},"Hochreiter and Schmidhuber (1997)",[94,13014,13015],{},"Long Short-Term Memory",", Neural Computation — the LSTM and its constant-error-carousel cell that keeps the recurrent gradient from vanishing or exploding; the forget gate was added by ",[21,13018,13019],{},"Gers, Schmidhuber, and Cummins (2000)",[94,13021,13022],{},"Learning to Forget",", Neural Computation. ",[15,13025,12962],{"href":13026,"ariaLabel":13027,"className":13028,"dataFootnoteBackref":6},"#user-content-fnref-hochreiter","Back to reference 6",[12961],[12759,13030,13032,1075,13035,13038,13039,1075,13042,13045,13046],{"id":13031},"user-content-fn-seq2seq",[21,13033,13034],{},"Sutskever, Vinyals, and Le (2014)",[94,13036,13037],{},"Sequence to Sequence Learning with Neural Networks",", NeurIPS — a deep LSTM encoder-decoder for machine translation, with source reversal; and ",[21,13040,13041],{},"Cho et al. (2014)",[94,13043,13044],{},"Learning Phrase Representations using RNN Encoder-Decoder",", EMNLP — the GRU and the RNN encoder-decoder. ",[15,13047,12962],{"href":13048,"ariaLabel":13049,"className":13050,"dataFootnoteBackref":6},"#user-content-fnref-seq2seq","Back to reference 7",[12961],[12759,13052,13054,1075,13057,13060,13061,1075,13064,13067,13068],{"id":13053},"user-content-fn-bahdanau",[21,13055,13056],{},"Bahdanau, Cho, and Bengio (2015)",[94,13058,13059],{},"Neural Machine Translation by Jointly Learning to Align and Translate",", ICLR — the attention mechanism that lets the decoder attend to all encoder states, removing the fixed-vector bottleneck; the recurrence-free ",[21,13062,13063],{},"Vaswani et al. (2017)",[94,13065,13066],{},"Attention Is All You Need",", NeurIPS, is the transformer. ",[15,13069,12962],{"href":13070,"ariaLabel":13071,"className":13072,"dataFootnoteBackref":6},"#user-content-fnref-bahdanau","Back to reference 8",[12961],[12759,13074,13076,13078,13079],{"id":13075},"user-content-fn-jm-seq2seq",[21,13077,12764],{},", Ch. 10 — The Encoder-Decoder Model: two RNNs joined by a context vector, the decoder as a conditioned autoregressive language model, and the fixed-length-context bottleneck that motivates attention. ",[15,13080,12962],{"href":13081,"ariaLabel":13082,"className":13083,"dataFootnoteBackref":6},"#user-content-fnref-jm-seq2seq","Back to reference 9",[12961],[13085,13086,13087],"style",{},"html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark-mode .shiki span {color: var(--shiki-dark-mode);background: var(--shiki-dark-mode-bg);font-style: var(--shiki-dark-mode-font-style);font-weight: var(--shiki-dark-mode-font-weight);text-decoration: var(--shiki-dark-mode-text-decoration);}html.dark-mode .shiki span {color: var(--shiki-dark-mode);background: var(--shiki-dark-mode-bg);font-style: var(--shiki-dark-mode-font-style);font-weight: var(--shiki-dark-mode-font-weight);text-decoration: var(--shiki-dark-mode-text-decoration);}",{"title":6,"searchDepth":2103,"depth":2103,"links":13089},[13090,13094,13097,13100,13101,13105,13106,13107,13108],{"id":129,"depth":2103,"text":130,"children":13091},[13092,13093],{"id":1965,"depth":2109,"text":1966},{"id":2142,"depth":2109,"text":2143},{"id":3388,"depth":2103,"text":3389,"children":13095},[13096],{"id":4960,"depth":2109,"text":4961},{"id":5721,"depth":2103,"text":5722,"children":13098},[13099],{"id":6023,"depth":2109,"text":6024},{"id":7886,"depth":2103,"text":7887},{"id":8060,"depth":2103,"text":8061,"children":13102},[13103,13104],{"id":10491,"depth":2109,"text":10492},{"id":11733,"depth":2109,"text":11734},{"id":11752,"depth":2103,"text":11753},{"id":11839,"depth":2103,"text":11840},{"id":12662,"depth":2103,"text":12663},{"id":123,"depth":2103,"text":12754},[],"computer-science","The feedforward neural language model\nfixed the sparsity of n-grams by embedding words as vectors, but it kept one limitation\nfrom them: a fixed window. It reads the last N−1 words, concatenates\ntheir embeddings, and predicts the next word — and a word N positions back is simply\ninvisible. Widen the window and the input layer grows with it; the model still cannot\ncondition on a subject that appeared twenty words ago. Linguistic dependencies span\narbitrary distances:\nthe verb in the keys to the cabinet are on the table agrees with keys, five words\nback, across a singular noun that would mislead any fixed-window guess.",false,"md",{"moduleNumber":2115,"lessonNumber":2109,"order":13115},403,"Sequences",true,"\u002Fnatural-language-processing\u002Fsequences\u002Frnns-and-lstms",[],"---\ntitle: RNNs and LSTMs\nmodule: Sequences\nmoduleNumber: 4\nlessonNumber: 3\norder: 403\nsummary: >\n  A feedforward neural language model sees a fixed window of words and can look no\n  further back. The recurrent neural network removes that limit: it carries a hidden\n  state across time, so each word is read in the context of everything before it. We\n  build the RNN from its one recurrent equation, use it as a language model, train it\n  by backpropagation through time, and diagnose the vanishing-gradient problem that\n  makes plain RNNs forget. The LSTM fixes the forgetting with a cell state and three\n  gates, and the encoder-decoder stacks two RNNs into a sequence-to-sequence model —\n  and its single-vector bottleneck is the problem attention was invented to remove.\ntopics: [Sequences]\nsources:\n  - book: Jurafsky\n    ref: \"Ch. 9 — Deep Learning Architectures for Sequence Processing; §9.2 Recurrent Neural Networks; §9.3 RNNs as Language Models\"\n  - book: Jurafsky\n    ref: \"§9.4 RNNs for other NLP tasks; §9.5 Stacked and Bidirectional RNNs; §9.6 The LSTM; Ch. 10 — The Encoder-Decoder Model\"\n---\n\nThe [feedforward neural language model](\u002Fnatural-language-processing\u002Fsemantics\u002Fneural-language-models)\nfixed the sparsity of n-grams by embedding words as vectors, but it kept one limitation\nfrom them: a **fixed window**. It reads the last $N-1$ words, concatenates\ntheir embeddings, and predicts the next word — and a word $N$ positions back is simply\ninvisible. Widen the window and the input layer grows with it; the model still cannot\ncondition on a subject that appeared twenty words ago. Linguistic dependencies span\narbitrary distances:\nthe verb in _the keys to the cabinet **are** on the table_ agrees with _keys_, five words\nback, across a singular noun that would mislead any fixed-window guess.\n\nThe **recurrent neural network** removes the window entirely. Instead of reading a fixed\nwindow of recent words, it processes the sequence one token at a time and carries a **hidden\nstate** forward from step to step. That state is a running summary of everything read so\nfar, with no built-in limit on how far back it can reach.[^jm-rnn] This lesson builds the\nRNN from its one recurrent equation, turns it into a language model, trains it, finds the\ngradient problem that undermines it, and repairs that problem with the LSTM.\n\n## The recurrent equation\n\nA recurrent network is any network with a cycle in its connections: the value of some unit\ndepends, directly or indirectly, on its own earlier output. The constrained, well-behaved\nversion used throughout language processing is the **Elman network**, or **simple recurrent\nnetwork**. At each time step $t$ it takes the current input vector $\\mathbf{x}_t$ and the\nhidden state $\\mathbf{h}_{t-1}$ from the previous step, and produces a new hidden state:\n\n$$\n\\mathbf{h}_t \\;=\\; g\\!\\left(\\mathbf{U}\\mathbf{h}_{t-1} + \\mathbf{W}\\mathbf{x}_t\\right),\n$$\n\nwhere $g$ is a non-linearity (typically $\\tanh$), $\\mathbf{W}$ maps the input into the\nhidden layer, and $\\mathbf{U}$ maps the previous hidden state into the current one. The\nrecurrent matrix $\\mathbf{U}$ is the whole difference from a feedforward network: it is how\npast context flows into the present. From the hidden state we read an output in the usual\nway, a softmax over classes for soft classification:\n\n$$\n\\mathbf{y}_t \\;=\\; \\mathrm{softmax}\\!\\left(\\mathbf{V}\\mathbf{h}_t\\right).\n$$\n\nThree weight matrices define the network — $\\mathbf{W} \\in \\mathbb{R}^{d_h \\times d_{in}}$,\n$\\mathbf{U} \\in \\mathbb{R}^{d_h \\times d_h}$, $\\mathbf{V} \\in \\mathbb{R}^{d_{out} \\times d_h}$\n— and, decisively, **they are shared across every time step**. The same $\\mathbf{U}$ that\ncarries step $1$'s context into step $2$ carries step $99$'s into step $100$. There is no\nper-position parameter and therefore no window; a sequence of any length is processed by the\nsame small set of weights.\n\n$$\n% caption: One RNN cell. The current input $\\mathbf{x}_t$ and the previous hidden state\n% $\\mathbf{h}_{t-1}$ combine through $\\mathbf{W}$ and $\\mathbf{U}$ into the new hidden state\n% $\\mathbf{h}_t$, which both feeds the output $\\mathbf{y}_t$ and is passed forward in time.\n\\begin{tikzpicture}[>=stealth, font=\\small,\n  box\u002F.style={draw, minimum width=13mm, minimum height=9mm, align=center}]\n  \\definecolor{acc}{HTML}{2348F2}\n  \\node[box] (x)  at (0,0)   {x(t)};\n  \\node[box, draw=acc, text=acc, thick] (h) at (0,1.9) {h(t)};\n  \\node[box] (y)  at (0,3.8) {y(t)};\n  \\draw[->, acc, thick] (x) -- (h) node[midway, right, font=\\scriptsize] {W};\n  \\draw[->, acc, thick] (h) -- (y) node[midway, right, font=\\scriptsize] {V};\n  % recurrent self-loop carrying h(t-1) into h(t)\n  \\draw[->, thick] (h.north west) .. controls (-1.8,3.1) and (-1.8,0.7) .. (h.south west);\n  \\node[font=\\scriptsize, anchor=east] at (-1.85,1.9) {U (from h(t-1))};\n\\end{tikzpicture}\n$$\n\nWritten this way the recurrence looks like new machinery, but it is not. If we **unroll** the\nnetwork in time — copy the cell once per time step and connect each copy's hidden state to the\nnext — the cycle disappears and what remains is an ordinary feedforward computation, just a\nvery deep one whose depth equals the sequence length. The copies are distinct in their\n_values_ (each $\\mathbf{h}_t$ differs) but identical in their _weights_ ($\\mathbf{U}$,\n$\\mathbf{V}$, $\\mathbf{W}$ are the same everywhere).\n\n$$\n% caption: The same RNN unrolled across three steps. Layers are recomputed at every step,\n% but the weights $\\mathbf{U}$, $\\mathbf{V}$, $\\mathbf{W}$ are shared; the hidden state is\n% the one wire that carries context from each step to the next.\n\\begin{tikzpicture}[>=stealth, font=\\small,\n  box\u002F.style={draw, minimum width=11mm, minimum height=8mm, align=center},\n  hbox\u002F.style={draw=acc, text=acc, thick, minimum width=11mm, minimum height=8mm, align=center}]\n  \\definecolor{acc}{HTML}{2348F2}\n  \\foreach \\t\u002F\\xx in {1\u002F0, 2\u002F3.3, 3\u002F6.6}{\n    \\node[box]  (x\\t) at (\\xx,0)   {x(\\t)};\n    \\node[hbox] (h\\t) at (\\xx,1.9) {h(\\t)};\n    \\node[box]  (y\\t) at (\\xx,3.8) {y(\\t)};\n    \\draw[->, acc, thick] (x\\t) -- (h\\t) node[midway, right, font=\\scriptsize] {W};\n    \\draw[->, acc, thick] (h\\t) -- (y\\t) node[midway, right, font=\\scriptsize] {V};\n  }\n  \\node[left=2mm of h1, font=\\scriptsize] {h(0)};\n  \\draw[->, thick] ($(h1.west)+(-1.3,0)$) -- (h1.west);\n  \\draw[->, thick] (h1) -- (h2) node[midway, above, font=\\scriptsize] {U};\n  \\draw[->, thick] (h2) -- (h3) node[midway, above, font=\\scriptsize] {U};\n\\end{tikzpicture}\n$$\n\n> **Definition (Recurrent neural network).** A network that computes a sequence of hidden\n> states $\\mathbf{h}_1, \\ldots, \\mathbf{h}_n$ by the recurrence\n> $\\mathbf{h}_t = g(\\mathbf{U}\\mathbf{h}_{t-1} + \\mathbf{W}\\mathbf{x}_t)$, sharing the\n> weights $\\mathbf{U}$, $\\mathbf{W}$ across all $t$. The state $\\mathbf{h}_t$ encodes the\n> whole prefix $\\mathbf{x}_1, \\ldots, \\mathbf{x}_t$, so context is not bounded by a fixed\n> window.\n\n### Forward inference\n\nBecause $\\mathbf{h}_t$ needs $\\mathbf{h}_{t-1}$, inference is inherently sequential: we sweep\nleft to right, and at each step compute the hidden state from the previous one and read off\nthe output. The whole forward pass is four lines.\n\n```algorithm\ncaption: $\\textsc{Forward-RNN}(\\mathbf{x}, \\mathit{network})$ — map an input sequence to outputs\ninput: sequence $\\mathbf{x} = \\mathbf{x}_1, \\ldots, \\mathbf{x}_n$; weights $\\mathbf{U}, \\mathbf{V}, \\mathbf{W}$\n$\\mathbf{h}_0 \\gets \\mathbf{0}$\nfor $t = 1$ to $n$ do\n  $\\mathbf{h}_t \\gets g(\\mathbf{U}\\mathbf{h}_{t-1} + \\mathbf{W}\\mathbf{x}_t)$ \u002F\u002F carry context forward\n  $\\mathbf{y}_t \\gets f(\\mathbf{V}\\mathbf{h}_t)$ \u002F\u002F read the output\nreturn $\\mathbf{y}_1, \\ldots, \\mathbf{y}_n$\n```\n\nThe sequential dependence is both the strength (unbounded context) and the weakness (no\nparallelism across time) of the architecture, and that lack of parallelism is what later\nmotivates the transformer.\n\n### Worked example: a two-step forward pass\n\nRun the recurrence with tiny numbers. Take one-dimensional hidden and input states,\n$g = \\tanh$, and weights $\\mathbf{W} = 0.8$, $\\mathbf{U} = 0.5$, with $h_0 = 0$. Feed\nthe inputs $x_1 = 1.0$ then $x_2 = -0.5$.\n\n**Step 1.** The pre-activation is $\\mathbf{U}h_0 + \\mathbf{W}x_1 = 0.5(0) + 0.8(1.0) =\n0.8$, so $h_1 = \\tanh(0.8) = 0.664$. The state now carries a trace\nof $x_1$.\n\n**Step 2.** The pre-activation is $\\mathbf{U}h_1 + \\mathbf{W}x_2 = 0.5(0.664) + 0.8(-0.5)\n= 0.332 - 0.400 = -0.068$, so $h_2 = \\tanh(-0.068) = -0.068$. Notice that $h_2$ depends\non $x_1$ _only through_ $h_1$ — the recurrent term $\\mathbf{U}h_1$ is the single channel\nby which step 1's information reaches step 2. This is the whole mechanism: everything the\npast contributes to the present flows through the one hidden wire, which is also why a\nweak recurrent weight (here $0.5$) attenuates old context, foreshadowing the vanishing\ngradient.\n\n## RNNs as language models\n\nTo use an RNN as a language model, run it over a corpus and, at each step, have it predict the\nnext word from the current word and the hidden state.[^jm-lm] Unlike the fixed-window neural\nmodel, the RNN has no context limit: $\\mathbf{h}_t$ can in principle carry information from the\nvery first word of the sequence.\n\nAt step $t$ the current word is looked up in the embedding matrix $\\mathbf{E}$, combined with\nthe previous hidden state, and turned into a distribution over the whole vocabulary:\n\n$$\n\\mathbf{e}_t \\;=\\; \\mathbf{E}\\mathbf{x}_t, \\qquad\n\\mathbf{h}_t \\;=\\; g\\!\\left(\\mathbf{U}\\mathbf{h}_{t-1} + \\mathbf{W}\\mathbf{e}_t\\right), \\qquad\n\\mathbf{y}_t \\;=\\; \\mathrm{softmax}\\!\\left(\\mathbf{V}\\mathbf{h}_t\\right).\n$$\n\nThe output $\\mathbf{y}_t$ is a probability vector over the vocabulary; its $i$-th component is\nthe model's estimate that word $i$ comes next, $P(w_{t+1} = i \\mid w_1, \\ldots, w_t) =\n\\mathbf{y}_t[i]$. The probability of a whole sequence is the product across positions,\n\n$$\nP(w_{1:n}) \\;=\\; \\prod_{i=1}^{n} P(w_i \\mid w_{1:i-1}) \\;=\\; \\prod_{i=1}^{n} \\mathbf{y}_i[w_i],\n$$\n\nwhere $\\mathbf{y}_i[w_i]$ denotes the probability the model placed on the _true_ word $w_i$ at\nstep $i$.\n\n### Training and teacher forcing\n\nThe RNN is trained to minimize the cross-entropy of the true next word. Because the target\ndistribution is a one-hot vector on the correct word $w_{t+1}$, the cross-entropy at step $t$\ncollapses to a single term, the negative log probability the model assigned to that word:\n\n$$\nL_{\\mathrm{CE}}(\\hat{\\mathbf{y}}_t, \\mathbf{y}_t) \\;=\\; -\\log \\hat{\\mathbf{y}}_t[w_{t+1}].\n$$\n\nThe loss for a training sequence is the average of these over all positions. One subtlety\ngoverns training: at each step we feed the model the **correct** history $w_{1:t}$, not the\nwords it actually predicted. If the model guesses wrong at step $t$, we ignore that guess and\nstill hand it the true $w_{1:t+1}$ for the next step. Always supplying the gold prefix is called\n**teacher forcing**, and it keeps training stable by preventing early mistakes from cascading.\n\n$$\n% caption: Training an RNN language model. Each step reads its input embedding, updates the\n% hidden state, and produces a softmax over the vocabulary; the loss is the negative log\n% probability of the true next word, averaged across the sequence.\n\\begin{tikzpicture}[>=stealth, font=\\small,\n  emb\u002F.style={draw, minimum width=9mm, minimum height=6mm, align=center, font=\\scriptsize},\n  rbox\u002F.style={draw=acc, text=acc, thick, minimum width=9mm, minimum height=8mm, align=center},\n  sm\u002F.style={draw, minimum width=9mm, minimum height=6mm, align=center, font=\\scriptsize},\n  ls\u002F.style={draw, minimum width=13mm, minimum height=6mm, align=center, font=\\scriptsize}]\n  \\definecolor{acc}{HTML}{2348F2}\n  \\def\\dx{2.7}\n  \\foreach \\i\u002F\\w\u002F\\nw in {0\u002FSo\u002Flong, 1\u002Flong\u002Fand, 2\u002Fand\u002Fthanks, 3\u002Fthanks\u002Ffor}{\n    \\node[emb] (e\\i) at (\\i*\\dx,0)   {\\w};\n    \\node[rbox] (h\\i) at (\\i*\\dx,1.5) {h};\n    \\node[sm]  (s\\i) at (\\i*\\dx,3.0) {softmax};\n    \\node[ls]  (l\\i) at (\\i*\\dx,4.4) {-log y(\\nw)};\n    \\draw[->, acc, thick] (e\\i) -- (h\\i);\n    \\draw[->, acc, thick] (h\\i) -- (s\\i);\n    \\draw[->, thick] (s\\i) -- (l\\i);\n  }\n  \\foreach \\i [evaluate=\\i as \\j using int(\\i+1)] in {0,1,2}{\n    \\draw[->, thick] (h\\i) -- (h\\j) node[midway, above, font=\\scriptsize] {h};\n  }\n  \\node[font=\\scriptsize, anchor=east] at (-0.8,0)   {input};\n  \\node[font=\\scriptsize, anchor=east] at (-0.8,1.5) {RNN};\n  \\node[font=\\scriptsize, anchor=east] at (-0.8,4.4) {loss};\n\\end{tikzpicture}\n$$\n\nThe same trained model can **generate** text: seed it with a start symbol, sample a word from\n$\\mathbf{y}_1$, feed that word back in as the next input, and repeat until an end symbol appears.\nBecause each generated word is fed back as the next step's input, this is called\n**autoregressive generation** — the model conditions on its own choices, one word at a time.\n\nRNNs slot into other tasks with only the output layer changed. For **sequence labeling** — the\n[part-of-speech and named-entity tagging](\u002Fnatural-language-processing\u002Fsequences\u002Fsequence-labeling)\nof the previous lesson — a softmax over the tagset at each step emits one label per token. For\n**sequence classification**, the final hidden state $\\mathbf{h}_n$ serves as a compressed\nsummary of the whole text and is fed to a feedforward classifier; the classification loss then\nbackpropagates all the way through the RNN, an instance of **end-to-end training**.\n\n## Backpropagation through time\n\nTraining an unrolled RNN is ordinary backpropagation applied to the unrolled graph, with one\nnew complication: a hidden state $\\mathbf{h}_t$ influences the loss at _every_ later step,\nbecause it feeds $\\mathbf{h}_{t+1}$, which feeds $\\mathbf{h}_{t+2}$, and so on. To find the\ngradient reaching $\\mathbf{h}_t$ we must sum the error flowing back from all of those futures.\nThe resulting two-pass algorithm — forward to compute and save every hidden state and loss,\nthen backward through time to accumulate gradients — is **backpropagation through time**\n(BPTT).[^jm-bptt]\n\nBecause the weights are shared, each of $\\mathbf{U}$, $\\mathbf{V}$, $\\mathbf{W}$ receives a\ngradient contribution from every time step; the per-step gradients are summed before the update.\nThe mechanics are standard, but the shape of the backward pass has a consequence that governs\nwhether the network can actually learn long-range structure.\n\n### Worked example: unrolling the backward chain\n\nUnroll three steps and follow one gradient path. The loss at step $3$, call it $L_3$,\ndepends on $\\mathbf{h}_3$, which depends on $\\mathbf{h}_2$, which depends on\n$\\mathbf{h}_1$. The chain rule for the gradient of $L_3$ with respect to the earliest\nhidden state $\\mathbf{h}_1$ is a product of per-step Jacobians:\n\n$$\n\\frac{\\partial L_3}{\\partial \\mathbf{h}_1} = \\frac{\\partial L_3}{\\partial \\mathbf{h}_3}\\;\n\\frac{\\partial \\mathbf{h}_3}{\\partial \\mathbf{h}_2}\\;\n\\frac{\\partial \\mathbf{h}_2}{\\partial \\mathbf{h}_1},\n\\qquad\n\\frac{\\partial \\mathbf{h}_t}{\\partial \\mathbf{h}_{t-1}} = \\mathrm{diag}\\!\\left(g'(\\mathbf{z}_t)\\right)\\mathbf{U},\n$$\n\nwhere $\\mathbf{z}_t = \\mathbf{U}\\mathbf{h}_{t-1} + \\mathbf{W}\\mathbf{x}_t$. Each hop back\nmultiplies by the _same_ factor $\\mathrm{diag}(g')\\mathbf{U}$, so crossing $k$ steps\nmultiplies by that factor $k$ times. For example, suppose the recurrent\nweight and the local derivative combine to a factor of $0.6$ per step. Then a gradient\ncrossing $1$ step is scaled by $0.6$, across $5$ steps by $0.6^5 = 0.078$, across $10$\nsteps by $0.6^{10} = 0.006$, and across $20$ steps by $0.6^{20} = 0.00004$. The signal\nreaching a state twenty words back is four orders of magnitude weaker than the one\nreaching the immediately previous state.\n\n$$\n% caption: The per-step multiplier compounds geometrically over $k$ steps back. With a\n% factor of $0.6$ per hop the gradient scale is $0.6^k$: it falls below $0.01$ by about\n% eight steps, so context more than a few words back gets a negligible learning signal.\n\\begin{tikzpicture}[>=stealth, font=\\footnotesize]\n  \\definecolor{acc}{HTML}{2348F2}\n  \\definecolor{red}{HTML}{C0392B}\n  \\draw[->, black] (0,0) -- (7.4,0) node[right, black, font=\\scriptsize] {steps back k};\n  \\draw[->, black] (0,0) -- (0,3.3) node[above, black, font=\\scriptsize] {gradient scale};\n  % plot 3*0.6^x for x=0..10 (scaled)\n  \\draw[acc, very thick, domain=0:10, samples=40, variable=\\k]\n    plot ({\\k*0.68}, {3*pow(0.6,\\k)});\n  \\node[acc, anchor=west] at (5.0,0.55) {0.6 to the k};\n  % markers\n  \\foreach \\k\u002F\\lab in {0\u002F1, 5\u002F5, 10\u002F10}\n    \\node[font=\\scriptsize, anchor=north] at (\\k*0.68,-0.05) {\\lab};\n  % 0.01 threshold line\n  \\draw[red, dashed] (0,0.09) -- (7.0,0.09);\n  \\node[red, anchor=south east, font=\\scriptsize] at (7.0,0.1) {below 0.01};\n\\end{tikzpicture}\n$$\n\nThe same product governs $\\partial L \u002F \\partial \\mathbf{U}$ and the other weight\ngradients, so a plain RNN cannot learn to connect a decision to context more than a\nhandful of steps back — precisely the failure the next section names.\n\n## The vanishing-gradient problem\n\nConsider how an error at step $t$ reaches a hidden state many steps earlier, at step\n$t - k$. The gradient must flow backward through the recurrence once per intervening step, and\neach such hop multiplies by the recurrent weight $\\mathbf{U}$ and by the derivative of the\nnon-linearity $g$. Over $k$ steps the signal is multiplied by roughly the same factor $k$\ntimes — a geometric progression.\n\nIf that repeated factor is less than one, the product shrinks toward zero: the gradient\n**vanishes**, and the far-past hidden states receive essentially no learning signal. If the\nfactor exceeds one, the product blows up and the gradient **explodes**. In practice, plain RNNs\nvanish far more often than they explode, and the effect is that the network cannot tie a\ndecision to context more than a handful of steps back.[^jm-vanish]\n\n$$\n% caption: The gradient's magnitude as it is propagated back through time, one recurrent hop\n% per step. Multiplying by a factor below one at each hop shrinks it geometrically, so distant\n% steps get almost no learning signal — the vanishing-gradient problem.\n\\begin{tikzpicture}[>=stealth, font=\\footnotesize]\n  \\definecolor{acc}{HTML}{2348F2}\n  \\definecolor{red}{HTML}{C0392B}\n  % axes\n  \\draw[->, black] (0,0) -- (8.4,0) node[right, black, font=\\scriptsize] {steps back};\n  \\draw[->, black] (0,0) -- (0,3.4) node[above, black, font=\\scriptsize] {gradient size};\n  % decaying gradient bars\n  \\foreach \\i\u002F\\hh in {1\u002F3.0, 2\u002F1.95, 3\u002F1.27, 4\u002F0.82, 5\u002F0.53, 6\u002F0.35, 7\u002F0.22}{\n    \\fill[acc!25, draw=acc] (\\i-0.28,0) rectangle (\\i+0.28,\\hh);\n  }\n  % decay curve just past the bar tops\n  \\draw[red, very thick]\n    (1,3.0) .. controls (2,2.1) and (3,1.3) .. (4,0.82)\n    .. controls (5,0.5) and (6,0.3) .. (7,0.22);\n  \\node[red, anchor=west] at (7.15,0.22) {vanishes};\n  % x tick labels\n  \\foreach \\i\u002F\\lab in {1\u002F1, 4\u002F4, 7\u002F7}\n    \\node[font=\\scriptsize, anchor=north] at (\\i,-0.05) {\\lab};\n\\end{tikzpicture}\n$$\n\nThe deeper diagnosis is that a plain RNN requires one hidden layer to do two conflicting jobs at\nonce: supply the information the _current_ step needs, and preserve information that some\n_future_ step will need. Every write for the present overwrites part of the past. Consider\n_the flights the airline was cancelling **were** full_: the number of _were_ depends on\n_flights_, but the intervening singular _airline_ dominates the local context and the plural\nsignal is written over long before it is needed. A network with a single, always-updated state\nhas no way to hold a value aside untouched.\n\n## The LSTM\n\nThe **long short-term memory** network solves this by giving the unit a second state — a\n**cell state** $\\mathbf{c}_t$ that runs alongside the hidden state and is engineered so that\ninformation can pass along it nearly unchanged.[^jm-lstm] The network then learns, through small\nsub-networks called **gates**, which information to erase from the cell, which to add, and which\nto expose as output. Context management is learned rather than fixed.\n\nEvery gate shares one design: a feedforward layer, a sigmoid $\\sigma$ squashing its output into\n$[0, 1]$, and a pointwise (Hadamard) multiplication $\\odot$ with the vector being gated. Because\nthe sigmoid pushes toward $0$ or $1$, the gate acts as a soft binary mask — coordinates near $1$\npass through, coordinates near $0$ are erased.\n\n**The forget gate** decides what to delete from the previous cell state. It reads the previous\nhidden state and current input, and its mask $\\mathbf{f}_t$ multiplies $\\mathbf{c}_{t-1}$\nelement-wise, zeroing out whatever is no longer relevant:\n\n$$\n\\mathbf{f}_t \\;=\\; \\sigma\\!\\left(\\mathbf{U}_f\\mathbf{h}_{t-1} + \\mathbf{W}_f\\mathbf{x}_t\\right),\n\\qquad\n\\mathbf{k}_t \\;=\\; \\mathbf{c}_{t-1} \\odot \\mathbf{f}_t.\n$$\n\n**The candidate** is the new information proposed for the cell — the same $\\tanh$ computation a\nplain RNN would perform:\n\n$$\n\\mathbf{g}_t \\;=\\; \\tanh\\!\\left(\\mathbf{U}_g\\mathbf{h}_{t-1} + \\mathbf{W}_g\\mathbf{x}_t\\right).\n$$\n\n**The input gate** selects which of that candidate to actually add, then the surviving old\ncontext and the selected new context are summed into the updated cell state:\n\n$$\n\\mathbf{i}_t \\;=\\; \\sigma\\!\\left(\\mathbf{U}_i\\mathbf{h}_{t-1} + \\mathbf{W}_i\\mathbf{x}_t\\right),\n\\qquad\n\\mathbf{c}_t \\;=\\; \\mathbf{k}_t + \\mathbf{g}_t \\odot \\mathbf{i}_t.\n$$\n\n**The output gate** decides what of the cell to expose as the hidden state used for the current\noutput and passed to the next step:\n\n$$\n\\mathbf{o}_t \\;=\\; \\sigma\\!\\left(\\mathbf{U}_o\\mathbf{h}_{t-1} + \\mathbf{W}_o\\mathbf{x}_t\\right),\n\\qquad\n\\mathbf{h}_t \\;=\\; \\mathbf{o}_t \\odot \\tanh(\\mathbf{c}_t).\n$$\n\n$$\n% caption: A single LSTM cell. The previous cell state $\\mathbf{c}_{t-1}$ flows along the top,\n% multiplied by the forget gate $f$ then added to the gated candidate ($g$ selected by the input\n% gate $i$); the output gate $o$ reads $\\tanh(\\mathbf{c}_t)$ into the new hidden state $\\mathbf{h}_t$.\n\\begin{tikzpicture}[>=stealth, font=\\footnotesize,\n  io\u002F.style={draw, minimum width=11mm, minimum height=6mm, align=center, font=\\scriptsize},\n  gate\u002F.style={draw, circle, minimum size=7mm, inner sep=0pt, font=\\scriptsize},\n  op\u002F.style={draw, circle, minimum size=6mm, inner sep=0pt, font=\\scriptsize},\n  tn\u002F.style={draw, minimum width=9mm, minimum height=5mm, align=center, font=\\scriptsize}]\n  \\definecolor{acc}{HTML}{2348F2}\n  % inputs on the left\n  \\node[io] (cprev) at (0,4.2) {c(t-1)};\n  \\node[io] (hprev) at (0,2.1) {h(t-1)};\n  \\node[io] (x)     at (0,0.4) {x(t)};\n  % cell-state line across the top\n  \\node[op] (fmul) at (3.2,4.2) {x};\n  \\node[op] (add)  at (6.4,4.2) {+};\n  \\node[io, draw=acc, text=acc, thick] (ct) at (9.4,4.2) {c(t)};\n  \\draw[->, thick] (cprev) -- (fmul);\n  \\draw[->, thick] (fmul) -- (add);\n  \\draw[->, acc, thick] (add) -- (ct);\n  % gates (well spaced, no overlaps)\n  \\node[gate] (f) at (3.2,2.4) {f};\n  \\node[gate] (g) at (5.0,1.0) {g};\n  \\node[gate] (i) at (5.0,2.5) {i};\n  \\node[op]   (gi) at (6.4,1.0) {x};\n  \\node[gate] (o) at (9.4,1.0) {o};\n  \\node[op]   (omul) at (9.4,2.6) {x};\n  \\node[tn]   (th)   at (7.9,3.3) {tanh};\n  \\node[io, draw=acc, text=acc, thick] (ht) at (11.4,1.0) {h(t)};\n  % gate feeds from h(t-1) and x(t) into forget gate\n  \\draw[->, black] (hprev) -- (f);\n  \\draw[->, black] (x) .. controls (1.4,0.4) and (2.2,1.4) .. (f.south);\n  \\draw[->, thick] (f) -- (fmul);\n  % candidate g gated by i, into the add (i sits off the vertical gi->add line)\n  \\draw[->, thick] (g) -- (gi);\n  \\draw[->, thick] (i) .. controls (5.6,2.3) and (6.0,1.6) .. (gi.north west);\n  \\draw[->, thick] (gi) -- (add);\n  % output path: c(t) -> tanh -> omul (gated by o) -> h(t)\n  \\draw[->, thick] (ct) .. controls (9.4,3.7) and (8.6,3.6) .. (th);\n  \\draw[->, thick] (th) .. controls (8.8,2.9) and (9.4,3.1) .. (omul);\n  \\draw[->, thick] (o) -- (omul);\n  \\draw[->, thick] (omul) .. controls (10.3,2.2) and (11.4,2.0) .. (ht);\n  \\draw[dashed, black] (-0.9,-0.2) rectangle (12.3,5.1);\n  \\node[anchor=south east, font=\\itshape, black] at (12.2,-0.15) {LSTM};\n\\end{tikzpicture}\n$$\n\nThe gate that matters most for the gradient is the forget gate on the cell state. The cell\nupdate $\\mathbf{c}_t = \\mathbf{c}_{t-1} \\odot \\mathbf{f}_t + \\ldots$ is (nearly) **additive**:\nwhen the forget gate stays open ($\\mathbf{f}_t \\approx 1$), backpropagating through the cell\nmultiplies the gradient by roughly $1$ at each step instead of by a shrinking factor. Context can\nsit in the cell across dozens of steps with its gradient intact — the additive path is\nwhat plain RNNs lacked, and the reason the LSTM can hold _flights_ until _were_ needs it.\n\n> **Definition (LSTM gate).** A feedforward layer followed by a sigmoid, whose output in\n> $[0,1]^d$ is multiplied element-wise into the vector it controls. The **forget gate** removes\n> information from the cell, the **input gate** admits new information, and the **output gate**\n> exposes the cell as the hidden state. Sigmoid gates turn context management into learned,\n> differentiable masking.\n\n### Worked example: the forget gate holds a value\n\nTake a single cell coordinate storing the \"subject is plural\" bit as $c = 1.0$, and\nwatch it survive intervening words. Suppose the network has learned, for this\ncoordinate, a forget gate that stays near open ($f = 0.95$) while it should keep the\nvalue, and near closed only when a new subject arrives.\n\n**A word that should not overwrite** (say the distractor _airline_). The forget gate\ncomputes $f = 0.95$, and the input gate computes $i = 0.05$ with candidate $g = -0.8$\n(a competing singular signal). The cell update is\n\n$$\nc_{\\text{new}} = f \\cdot c_{\\text{old}} + i \\cdot g = 0.95(1.0) + 0.05(-0.8) = 0.95 - 0.04 = 0.91.\n$$\n\nThe plural bit dropped only from $1.0$ to $0.91$ — the distractor barely dented it,\nbecause the input gate admitted little of the new candidate. Repeat this across five\nintervening words with the same gates and the value decays as $0.95^5 \\approx 0.77$,\nstill clearly positive when the verb finally needs it. Contrast a plain RNN, whose\nsingle state would be _overwritten_ by each word: with the tanh update of the earlier\nexample, the plural signal is gone within two or three steps. The forget gate near $1$\nis what turns the multiplicative decay into near-preservation.\n\n**The gradient view.** Because $c_t = f_t \\odot c_{t-1} + \\ldots$, the backward\nderivative $\\partial c_t \u002F \\partial c_{t-1} = f_t$. When $f_t \\approx 1$ the gradient\ncrossing the cell is multiplied by $\\approx 1$ per step instead of by the $\\approx 0.6$\nof the plain-RNN example — so $1^{20} = 1$ where $0.6^{20} = 0.00004$. The cell is a\ngradient-preserving path, and the forget gate controls it.\n\nThe full derivation of the gates, their gradients, and the additive cell-state path is in the\ndeep-learning course's [LSTM & GRU](\u002Fdeep-learning\u002Farchitectures\u002Flstm-and-gru) lesson; the RNN\nrecurrence and BPTT are treated in\n[recurrent networks](\u002Fdeep-learning\u002Farchitectures\u002Frecurrent-networks).\n\n### The GRU\n\nThe LSTM's three gates and separate cell state are effective but heavy — four feedforward layers\nper unit. The **gated recurrent unit** (GRU) is a lighter design that keeps the same\ngradient-preserving idea with fewer parts: it drops the separate cell state and uses just two\ngates, a _reset_ gate that controls how much past state enters the candidate and an _update_ gate\nthat interpolates between the old state and the candidate. It trains faster and often matches the\nLSTM's accuracy, so it is a common default when the LSTM's full capacity is not needed.\n\n## The LSTM lineage and the rise of attention\n\nThe recurrent architectures of this lesson have a precise public history, and the last\nstep in it is the one that made them obsolete for translation.\n\n**The LSTM** (Hochreiter and Schmidhuber, 1997).[^hochreiter] The long short-term\nmemory unit was introduced two decades before it became standard, specifically to solve\nthe vanishing-gradient problem this lesson diagnoses. Their analysis showed that a plain\nrecurrent unit's error signal decays or explodes exponentially in the time lag, and\ntheir fix — the \"constant error carousel,\" an internal cell whose self-connection has\nweight $1$ so the gradient neither grows nor shrinks — coincides with the additive cell\npath above. The forget gate was added later (Gers, Schmidhuber, and Cummins, 2000) and is now\npart of the standard cell.\n\n**Sequence-to-sequence and the GRU** (Sutskever, Vinyals, and Le, 2014; Cho et al.,\n2014).[^seq2seq] Two 2014 papers turned stacked LSTMs into machine-translation systems.\nSutskever and colleagues trained a deep LSTM encoder-decoder end to end on English-French\nand reached a BLEU score competitive with phrase-based statistical systems; one trick\ncarried much of the gain — _reversing_ the source sentence so early source words sat near the decoder.\nCho and colleagues introduced the GRU in the same encoder-decoder setting. Both hit the\nsingle-context-vector bottleneck this lesson closes on.\n\n**Attention** (Bahdanau, Cho, and Bengio, 2015).[^bahdanau] The fix for the bottleneck\nwas to let the decoder read a _weighted combination of all_ encoder hidden states at each\nstep, with the weights computed from how well each source position matches the current\ndecoder state. Bahdanau and colleagues showed this \"align and translate\" mechanism\nremoved the sharp quality drop on long sentences that the fixed-vector encoder-decoder\nsuffered. Attention was first an add-on to an RNN; the [transformer](\u002Fnatural-language-processing\u002Ftransformers\u002Ftransformers-and-attention)\n(Vaswani et al., 2017) then kept the attention and discarded the recurrence entirely,\nregaining the parallelism across time that the sequential RNN gave up — the exact\nweakness noted after the forward-inference algorithm above.\n\n## Stacked and bidirectional RNNs\n\nThe RNN cell is a module — vectors in, vectors out — so it composes in two useful ways.\n\nA **stacked RNN** feeds the entire output sequence of one RNN as the input sequence of another,\npiling several layers deep. Lower layers induce local, concrete features; higher layers build\nmore abstract representations on top, the same layered-abstraction argument that motivates depth\nelsewhere. Stacked RNNs generally outperform single-layer ones, though training cost rises with\ndepth.\n\nA **bidirectional RNN** addresses a different limitation: a left-to-right RNN's state at step $t$\nknows only the words up to $t$, nothing to the right. For tasks where the whole input is\navailable — labeling a token given its full sentence, classifying a complete document — the words\nafter $t$ are informative too. A bidirectional RNN runs two independent RNNs, one forward and one\nbackward over the reversed sequence, and concatenates their states at each position:\n\n$$\n\\mathbf{h}_t^{f} = \\mathrm{RNN}_{\\text{forward}}(\\mathbf{x}_1, \\ldots, \\mathbf{x}_t), \\qquad\n\\mathbf{h}_t^{b} = \\mathrm{RNN}_{\\text{backward}}(\\mathbf{x}_t, \\ldots, \\mathbf{x}_n), \\qquad\n\\mathbf{h}_t = [\\mathbf{h}_t^{f} \\oplus \\mathbf{h}_t^{b}].\n$$\n\nThe combined $\\mathbf{h}_t$ sees both the left and right context of position $t$ — the\nview a local labeling decision wants.\n\n$$\n% caption: A bidirectional RNN. A forward RNN and a backward RNN process the sequence in opposite\n% directions; their hidden states are concatenated at each position so every output sees both the\n% left and the right context.\n\\begin{tikzpicture}[>=stealth, font=\\footnotesize,\n  box\u002F.style={draw, minimum width=8mm, minimum height=7mm, align=center, font=\\scriptsize},\n  ob\u002F.style={draw, minimum width=8mm, minimum height=6mm, align=center, font=\\scriptsize}]\n  \\definecolor{acc}{HTML}{2348F2}\n  \\def\\dx{2.4}\n  \\foreach \\i in {1,2,3}{\n    \\node[box] (x\\i) at (\\i*\\dx,0)   {x(\\i)};\n    \\node[box] (f\\i) at (\\i*\\dx,1.5) {f};\n    \\node[box] (b\\i) at (\\i*\\dx,3.0) {b};\n    \\node[ob]  (y\\i) at (\\i*\\dx,4.5) {y(\\i)};\n    \\draw[->, thick] (x\\i) -- (f\\i);\n    \\draw[->, thick] (f\\i) -- (y\\i);\n    \\draw[->, thick] (b\\i) -- (y\\i);\n  }\n  % forward chain L->R\n  \\draw[->, acc, thick] (f1) -- (f2);\n  \\draw[->, acc, thick] (f2) -- (f3);\n  % backward chain R->L\n  \\draw[->, thick] (b3) -- (b2);\n  \\draw[->, thick] (b2) -- (b1);\n  \\node[acc, font=\\scriptsize, anchor=west] at (3*\\dx+0.35,1.5) {forward pass};\n  \\node[font=\\scriptsize, anchor=west] at (3*\\dx+0.35,3.0) {backward pass};\n\\end{tikzpicture}\n$$\n\n## The encoder-decoder model\n\nThe tasks so far map an input sequence to an output of the _same length_ — one tag per word, one\nlabel per document. Machine translation, summarization, and question answering are different: the\noutput is a new sequence whose length and word order need not match the input at all. The\narchitecture built for this is the **encoder-decoder**, or **sequence-to-sequence**, model.[^jm-seq2seq]\n\nIt is two RNNs joined at a single vector. The **encoder** reads the entire source sequence and\ncompresses it into a fixed-length **context vector** — typically its final hidden state. The\n**decoder** is an autoregressive RNN language model whose initial state is that context vector; it\ngenerates the target sequence one token at a time, each step conditioned on the context and the\nwords produced so far, until it emits an end symbol.\n\n$$\n% caption: The encoder-decoder model. The encoder RNN compresses the whole source into a single\n% context vector $\\mathbf{c}$, which seeds the decoder RNN; the decoder generates the target\n% autoregressively. Everything the decoder knows about the source must pass through $\\mathbf{c}$.\n\\begin{tikzpicture}[>=stealth, font=\\footnotesize,\n  box\u002F.style={draw, minimum width=10mm, minimum height=7mm, align=center, font=\\scriptsize},\n  cb\u002F.style={draw=acc, text=acc, thick, minimum width=11mm, minimum height=7mm, align=center, font=\\scriptsize}]\n  \\definecolor{acc}{HTML}{2348F2}\n  \\def\\dx{2.0}\n  % encoder\n  \\foreach \\i\u002F\\w in {1\u002Fle, 2\u002Fchat, 3\u002Fdort}{\n    \\node[box] (e\\i) at (\\i*\\dx,0) {\\w};\n    \\node[box] (E\\i) at (\\i*\\dx,1.5) {enc};\n    \\draw[->, thick] (e\\i) -- (E\\i);\n  }\n  \\draw[->, acc, thick] (E1) -- (E2);\n  \\draw[->, acc, thick] (E2) -- (E3);\n  % context vector, set apart from the last encoder box\n  \\node[cb] (c) at (3*\\dx+2.2,1.5) {c};\n  \\draw[->, acc, thick] (E3) -- (c) node[midway, above, font=\\scriptsize] {context};\n  % decoder (start symbol as ASCII text, not a bracketed token)\n  \\foreach \\i\u002F\\w\u002F\\nw in {1\u002Fstart\u002Fthe, 2\u002Fthe\u002Fcat, 3\u002Fcat\u002Fsleeps}{\n    \\node[box] (D\\i) at (3*\\dx+2.2+\\i*\\dx,1.5) {dec};\n    \\node[box] (d\\i) at (3*\\dx+2.2+\\i*\\dx,0)   {\\w};\n    \\node[box] (o\\i) at (3*\\dx+2.2+\\i*\\dx,3.0) {\\nw};\n    \\draw[->, thick] (d\\i) -- (D\\i);\n    \\draw[->, thick] (D\\i) -- (o\\i);\n  }\n  \\draw[->, acc, thick] (c) -- (D1);\n  \\draw[->, thick] (D1) -- (D2);\n  \\draw[->, thick] (D2) -- (D3);\n  \\node[font=\\scriptsize, anchor=north] at (2*\\dx,-0.8) {encoder (source)};\n  \\node[font=\\scriptsize, anchor=north] at (3*\\dx+2.2+2*\\dx,-0.8) {decoder (target)};\n\\end{tikzpicture}\n$$\n\nThis design trains end-to-end on paired sequences and produced the first strong neural machine\ntranslation systems. But it has a structural flaw visible in the figure: **everything the decoder\nlearns about the source must squeeze through the single context vector $\\mathbf{c}$**. For a short\nsentence that is fine; for a long paragraph, one fixed-length vector cannot hold every detail, and\ntranslation quality falls as the source grows. The information **bottleneck** is inherent to\ncompressing an arbitrarily long input into one vector.\n\nThe fix is to let the decoder look back at _all_ of the encoder's hidden states, not just the final\none, weighting them by relevance at each decoding step — the mechanism called **attention**. Attention\nremoves the bottleneck, and once a model is built entirely out of attention with the recurrence\nremoved, it becomes the [transformer](\u002Fnatural-language-processing\u002Ftransformers\u002Ftransformers-and-attention),\nthe subject of the next module.\n\n[^jm-rnn]: **Jurafsky & Martin**, _Speech and Language Processing_ (3rd ed.), §9.2 — Recurrent Neural Networks: the Elman\u002Fsimple recurrent network, the recurrence $\\mathbf{h}_t = g(\\mathbf{U}\\mathbf{h}_{t-1} + \\mathbf{W}\\mathbf{x}_t)$, weight sharing across time, and unrolling the cycle into a feedforward graph.\n[^jm-lm]: **Jurafsky & Martin**, §9.3 — RNNs as Language Models: next-word prediction with no fixed context window, the embedding-hidden-softmax forward pass, sequence probability as a product of per-step probabilities, and the cross-entropy training objective.\n[^jm-bptt]: **Jurafsky & Martin**, §9.2.2 — Training: the two-pass backpropagation-through-time algorithm, saving hidden states forward and accumulating gradients backward through time, with the three weight matrices shared across steps.\n[^jm-vanish]: **Jurafsky & Martin**, §9.6 — The LSTM (motivation): the difficulty of carrying distant information, the repeated multiplications during the backward pass that drive gradients to zero, and the vanishing-gradient problem.\n[^jm-lstm]: **Jurafsky & Martin**, §9.6 — The LSTM: the explicit cell\u002Fcontext layer, the forget, input (add), and output gates as sigmoid-masked feedforward layers, and the additive cell update that lets gradients flow across many steps.\n[^jm-seq2seq]: **Jurafsky & Martin**, Ch. 10 — The Encoder-Decoder Model: two RNNs joined by a context vector, the decoder as a conditioned autoregressive language model, and the fixed-length-context bottleneck that motivates attention.\n[^hochreiter]: **Hochreiter and Schmidhuber (1997)**, _Long Short-Term Memory_, Neural Computation — the LSTM and its constant-error-carousel cell that keeps the recurrent gradient from vanishing or exploding; the forget gate was added by **Gers, Schmidhuber, and Cummins (2000)**, _Learning to Forget_, Neural Computation.\n[^seq2seq]: **Sutskever, Vinyals, and Le (2014)**, _Sequence to Sequence Learning with Neural Networks_, NeurIPS — a deep LSTM encoder-decoder for machine translation, with source reversal; and **Cho et al. (2014)**, _Learning Phrase Representations using RNN Encoder-Decoder_, EMNLP — the GRU and the RNN encoder-decoder.\n[^bahdanau]: **Bahdanau, Cho, and Bengio (2015)**, _Neural Machine Translation by Jointly Learning to Align and Translate_, ICLR — the attention mechanism that lets the decoder attend to all encoder states, removing the fixed-vector bottleneck; the recurrence-free **Vaswani et al. (2017)**, _Attention Is All You Need_, NeurIPS, is the transformer.\n",{"text":13122,"minutes":13123,"time":13124,"words":13125},"17 min read",16.98,1018800,3396,{"title":5,"description":13111},[13128,13131],{"book":13129,"ref":13130},"Jurafsky","Ch. 9 — Deep Learning Architectures for Sequence Processing; §9.2 Recurrent Neural Networks; §9.3 RNNs as Language Models",{"book":13129,"ref":13132},"§9.4 RNNs for other NLP tasks; §9.5 Stacked and Bidirectional RNNs; §9.6 The LSTM; Ch. 10 — The Encoder-Decoder Model","available","09.natural-language-processing\u002F04.sequences\u002F03.rnns-and-lstms","A feedforward neural language model sees a fixed window of words and can look no further back. The recurrent neural network removes that limit: it carries a hidden state across time, so each word is read in the context of everything before it. We build the RNN from its one recurrent equation, use it as a language model, train it by backpropagation through time, and diagnose the vanishing-gradient problem that makes plain RNNs forget. The LSTM fixes the forgetting with a cell state and three gates, and the encoder-decoder stacks two RNNs into a sequence-to-sequence model — and its single-vector bottleneck is the problem attention was invented to remove.\n",[13116],"mTV6jNUAEyerSor4ewYHfH5zpBuVLkh1zanMGr1MIrM",{"\u002Falgorithms\u002Ffoundations\u002Fwhat-is-an-algorithm":13139,"\u002Falgorithms\u002Ffoundations\u002Fproof-techniques":13140,"\u002Falgorithms\u002Ffoundations\u002Fasymptotic-analysis":13141,"\u002Falgorithms\u002Ffoundations\u002Fgrowth-rates-and-loop-analysis":13142,"\u002Falgorithms\u002Ffoundations\u002Frecurrences":13143,"\u002Falgorithms\u002Ffoundations\u002Famortized-analysis":13144,"\u002Falgorithms\u002Fdivide-and-conquer\u002Fmergesort":13145,"\u002Falgorithms\u002Fdivide-and-conquer\u002Fquicksort":13146,"\u002Falgorithms\u002Fdivide-and-conquer\u002Fselection":13147,"\u002Falgorithms\u002Fdivide-and-conquer\u002Ffast-multiplication":13148,"\u002Falgorithms\u002Fsorting\u002Fheaps-and-heapsort":13149,"\u002Falgorithms\u002Fsorting\u002Fsorting-lower-bounds":13150,"\u002Falgorithms\u002Fsorting\u002Flinear-time-sorting":13151,"\u002Falgorithms\u002Fsorting\u002Fexternal-sorting":13152,"\u002Falgorithms\u002Fdata-structures\u002Felementary-structures":13153,"\u002Falgorithms\u002Fdata-structures\u002Fhash-tables":13154,"\u002Falgorithms\u002Fdata-structures\u002Fbinary-search-trees":13155,"\u002Falgorithms\u002Fdata-structures\u002Favl-trees":13156,"\u002Falgorithms\u002Fdata-structures\u002Fbalanced-trees":13157,"\u002Falgorithms\u002Fdata-structures\u002Funion-find":13158,"\u002Falgorithms\u002Fdata-structures\u002Ffenwick-and-segment-trees":13159,"\u002Falgorithms\u002Fdata-structures\u002Fspatial-data-structures":13160,"\u002Falgorithms\u002Fdata-structures\u002Fskip-lists-and-probabilistic-structures":13161,"\u002Falgorithms\u002Fdata-structures\u002Fb-trees":13162,"\u002Falgorithms\u002Fdata-structures\u002Fdata-stream-algorithms":13163,"\u002Falgorithms\u002Fdata-structures\u002Fstreaming-sketches":13164,"\u002Falgorithms\u002Fsequences\u002Ftwo-pointers-and-windows":13165,"\u002Falgorithms\u002Fsequences\u002Fprefix-sums":13166,"\u002Falgorithms\u002Fsequences\u002Fmonotonic-stacks":13167,"\u002Falgorithms\u002Fsequences\u002Fbinary-search-on-the-answer":13168,"\u002Falgorithms\u002Fsequences\u002Fstring-matching":13169,"\u002Falgorithms\u002Fsequences\u002Fkmp-and-z-function":13170,"\u002Falgorithms\u002Fsequences\u002Ftries":13171,"\u002Falgorithms\u002Fsequences\u002Fsuffix-arrays-and-aho-corasick":13172,"\u002Falgorithms\u002Fgraphs\u002Frepresentations-and-traversal":13173,"\u002Falgorithms\u002Fgraphs\u002Fdepth-first-search":13174,"\u002Falgorithms\u002Fgraphs\u002Ftopological-sort-and-scc":13175,"\u002Falgorithms\u002Fgraphs\u002Fminimum-spanning-trees":13176,"\u002Falgorithms\u002Fgraphs\u002Fkruskal-and-prim":13177,"\u002Falgorithms\u002Fgraphs\u002Fshortest-paths":13178,"\u002Falgorithms\u002Fgraphs\u002Fall-pairs-and-negative-weights":13179,"\u002Falgorithms\u002Fgraphs\u002Fnetwork-flow":13180,"\u002Falgorithms\u002Fgraphs\u002Fmax-flow-min-cut":13181,"\u002Falgorithms\u002Fgraphs\u002Fbridges-and-articulation-points":13182,"\u002Falgorithms\u002Fgraphs\u002Flowest-common-ancestor":13183,"\u002Falgorithms\u002Fgraphs\u002Ftwo-sat":13184,"\u002Falgorithms\u002Fgraphs\u002Feulerian-tours":13185,"\u002Falgorithms\u002Fgraphs\u002Fbipartite-matching":13186,"\u002Falgorithms\u002Fgreedy\u002Fthe-greedy-method":13187,"\u002Falgorithms\u002Fgreedy\u002Fscheduling-and-intervals":13188,"\u002Falgorithms\u002Fgreedy\u002Fhuffman-codes":13189,"\u002Falgorithms\u002Fgreedy\u002Fmatroids":13190,"\u002Falgorithms\u002Fgreedy\u002Fstable-matching":13191,"\u002Falgorithms\u002Fdynamic-programming\u002Fprinciples":13192,"\u002Falgorithms\u002Fdynamic-programming\u002Fsequence-dp":13193,"\u002Falgorithms\u002Fdynamic-programming\u002Flongest-increasing-subsequence":13194,"\u002Falgorithms\u002Fdynamic-programming\u002Fknapsack":13195,"\u002Falgorithms\u002Fdynamic-programming\u002Fcoin-change-and-unbounded":13196,"\u002Falgorithms\u002Fdynamic-programming\u002Finterval-dp":13197,"\u002Falgorithms\u002Fdynamic-programming\u002Ftree-dp":13198,"\u002Falgorithms\u002Fdynamic-programming\u002Fbitmask-dp":13199,"\u002Falgorithms\u002Fdynamic-programming\u002Fdp-optimizations":13200,"\u002Falgorithms\u002Fdynamic-programming\u002Fdp-on-graphs":13201,"\u002Falgorithms\u002Fdynamic-programming\u002Fdigit-and-probability-dp":13202,"\u002Falgorithms\u002Fbacktracking\u002Fbacktracking-fundamentals":13203,"\u002Falgorithms\u002Fbacktracking\u002Fconstraint-search":13204,"\u002Falgorithms\u002Fbacktracking\u002Fbranch-and-bound":13205,"\u002Falgorithms\u002Fbacktracking\u002Fgraph-backtracking":13206,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fnumber-theory-basics":13207,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fmodular-exponentiation-and-primality":13208,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fsieve-and-factorization":13209,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fcombinatorics":13210,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fmatrix-exponentiation":13211,"\u002Falgorithms\u002Fmathematical-algorithms\u002Ffast-fourier-transform":13212,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fgradient-descent":13213,"\u002Falgorithms\u002Fcomputational-geometry\u002Fgeometric-primitives":13214,"\u002Falgorithms\u002Fcomputational-geometry\u002Fconvex-hull":13215,"\u002Falgorithms\u002Fcomputational-geometry\u002Fsweep-line":13216,"\u002Falgorithms\u002Fcomputational-geometry\u002Fpolygons-and-proximity":13217,"\u002Falgorithms\u002Fintractability\u002Fp-np-reductions":13218,"\u002Falgorithms\u002Fintractability\u002Fnp-completeness":13219,"\u002Falgorithms\u002Fintractability\u002Fcoping-with-hardness":13220,"\u002Falgorithms\u002Fintractability\u002Fapproximation-algorithms":13221,"\u002Falgorithms":13222,"\u002Fcalculus\u002Flimits-and-continuity\u002Ffunctions-and-models":13223,"\u002Fcalculus\u002Flimits-and-continuity\u002Fthe-limit-of-a-function":13224,"\u002Fcalculus\u002Flimits-and-continuity\u002Flimit-laws-and-the-precise-definition":13225,"\u002Fcalculus\u002Flimits-and-continuity\u002Fcontinuity":13226,"\u002Fcalculus\u002Fderivatives\u002Fthe-derivative-and-rates-of-change":13227,"\u002Fcalculus\u002Fderivatives\u002Fdifferentiation-rules-and-the-chain-rule":13228,"\u002Fcalculus\u002Fderivatives\u002Fimplicit-differentiation-and-related-rates":13229,"\u002Fcalculus\u002Fderivatives\u002Flinear-approximations-and-differentials":13230,"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fextrema-and-the-mean-value-theorem":13231,"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fhow-derivatives-shape-a-graph":13232,"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fcurve-sketching-and-optimization":13233,"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fnewtons-method-and-antiderivatives":13234,"\u002Fcalculus\u002Fintegrals\u002Farea-and-the-definite-integral":13235,"\u002Fcalculus\u002Fintegrals\u002Fthe-fundamental-theorem-of-calculus":13236,"\u002Fcalculus\u002Fintegrals\u002Fthe-substitution-rule":13237,"\u002Fcalculus\u002Fapplications-of-integration\u002Fareas-and-volumes":13238,"\u002Fcalculus\u002Fapplications-of-integration\u002Fwork-average-value-and-arc-length":13239,"\u002Fcalculus\u002Fapplications-of-integration\u002Fphysics-economics-and-probability":13240,"\u002Fcalculus\u002Fexponential-logarithmic-and-inverse-functions\u002Finverse-functions-logarithms-and-exponentials":13241,"\u002Fcalculus\u002Fexponential-logarithmic-and-inverse-functions\u002Fgrowth-decay-inverse-trig-and-hyperbolic-functions":13242,"\u002Fcalculus\u002Fexponential-logarithmic-and-inverse-functions\u002Flhospitals-rule":13243,"\u002Fcalculus\u002Ftechniques-of-integration\u002Fintegration-by-parts":13244,"\u002Fcalculus\u002Ftechniques-of-integration\u002Ftrigonometric-integrals-and-substitution":13245,"\u002Fcalculus\u002Ftechniques-of-integration\u002Fpartial-fractions-and-integration-strategy":13246,"\u002Fcalculus\u002Ftechniques-of-integration\u002Fapproximate-and-improper-integrals":13247,"\u002Fcalculus\u002Fparametric-and-polar\u002Fparametric-curves-and-their-calculus":13248,"\u002Fcalculus\u002Fparametric-and-polar\u002Fpolar-coordinates":13249,"\u002Fcalculus\u002Fparametric-and-polar\u002Fconic-sections":13250,"\u002Fcalculus\u002Fsequences-and-series\u002Fsequences":13251,"\u002Fcalculus\u002Fsequences-and-series\u002Fseries-and-the-integral-test":13252,"\u002Fcalculus\u002Fsequences-and-series\u002Fthe-convergence-tests":13253,"\u002Fcalculus\u002Fsequences-and-series\u002Fpower-series":13254,"\u002Fcalculus\u002Fsequences-and-series\u002Ftaylor-and-maclaurin-series":13255,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fvectors-and-the-dot-product":13256,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fthe-cross-product-lines-and-planes":13237,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fcylinders-and-quadric-surfaces":13257,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fvector-functions-and-space-curves":13258,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Farc-length-curvature-and-motion":13259,"\u002Fcalculus\u002Fpartial-derivatives\u002Ffunctions-of-several-variables":13227,"\u002Fcalculus\u002Fpartial-derivatives\u002Fpartial-derivatives":13260,"\u002Fcalculus\u002Fpartial-derivatives\u002Ftangent-planes-and-the-chain-rule":13261,"\u002Fcalculus\u002Fpartial-derivatives\u002Fdirectional-derivatives-and-the-gradient":13262,"\u002Fcalculus\u002Fpartial-derivatives\u002Foptimization-and-lagrange-multipliers":13263,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fdouble-integrals":13264,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Ftriple-integrals-and-coordinate-systems":13265,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fvector-fields-and-line-integrals":13266,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fgreens-theorem-curl-and-divergence":13267,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fsurface-integrals":13268,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fstokes-and-the-divergence-theorem":13269,"\u002Fcalculus":13270,"\u002Fmechanics\u002Ffoundations\u002Fmeasurement-and-dimensions":13271,"\u002Fmechanics\u002Ffoundations\u002Fvector-algebra":13272,"\u002Fmechanics\u002Fkinematics\u002Fone-dimensional-motion":13273,"\u002Fmechanics\u002Fkinematics\u002Fmotion-graphs":13274,"\u002Fmechanics\u002Fkinematics\u002Fprojectile-motion":13275,"\u002Fmechanics\u002Fkinematics\u002Frelative-motion":13276,"\u002Fmechanics\u002Fkinematics\u002Fcircular-motion":13277,"\u002Fmechanics\u002Fdynamics\u002Fnewtons-laws":13278,"\u002Fmechanics\u002Fdynamics\u002Ffree-body-diagrams":13279,"\u002Fmechanics\u002Fdynamics\u002Ffriction-and-curved-motion":13280,"\u002Fmechanics\u002Fdynamics\u002Fnumerical-dynamics":13281,"\u002Fmechanics\u002Fdynamics\u002Fcenter-of-mass-systems":13282,"\u002Fmechanics\u002Fenergy\u002Fwork-and-kinetic-energy":13283,"\u002Fmechanics\u002Fenergy\u002Fpotential-energy":13284,"\u002Fmechanics\u002Fenergy\u002Fmultiparticle-work":13285,"\u002Fmechanics\u002Fenergy\u002Fmass-energy-and-binding":13286,"\u002Fmechanics\u002Fenergy\u002Fphotons-and-quantization":13287,"\u002Fmechanics\u002Fmomentum\u002Fmomentum-and-collisions":13288,"\u002Fmechanics\u002Fmomentum\u002Fcenter-of-mass-collisions":13289,"\u002Fmechanics\u002Fmomentum\u002Frocket-propulsion":13290,"\u002Fmechanics\u002Frotation\u002Frotational-inertia":13291,"\u002Fmechanics\u002Frotation\u002Frotational-dynamics":13292,"\u002Fmechanics\u002Frotation\u002Frolling-motion":13293,"\u002Fmechanics\u002Frotation\u002Fangular-momentum":13294,"\u002Fmechanics\u002Frotation\u002Frolling-resistance":13295,"\u002Fmechanics\u002Frotation\u002Fgyroscopic-precession":13296,"\u002Fmechanics\u002Fgravity-and-matter\u002Fkeplerian-orbits":13297,"\u002Fmechanics\u002Fgravity-and-matter\u002Fgravitational-fields":13298,"\u002Fmechanics\u002Fgravity-and-matter\u002Fstatic-equilibrium":13299,"\u002Fmechanics\u002Fgravity-and-matter\u002Ffluid-statics":13300,"\u002Fmechanics\u002Fgravity-and-matter\u002Ffluid-flow":13301,"\u002Fmechanics\u002Fgravity-and-matter\u002Forbital-motion":13302,"\u002Fmechanics\u002Fgravity-and-matter\u002Fstress-and-elasticity":13303,"\u002Fmechanics\u002Foscillations-waves\u002Fdamped-oscillators":13304,"\u002Fmechanics\u002Foscillations-waves\u002Ftravelling-waves":13305,"\u002Fmechanics\u002Foscillations-waves\u002Fwave-superposition":13306,"\u002Fmechanics\u002Foscillations-waves\u002Fstanding-waves":13307,"\u002Fmechanics\u002Foscillations-waves\u002Fsound-waves":13308,"\u002Fmechanics\u002Foscillations-waves\u002Fdoppler-effect":13309,"\u002Fmechanics\u002Foscillations-waves\u002Fwave-packets":13310,"\u002Fmechanics\u002Foscillations-waves\u002Fbeats-and-coupling":13311,"\u002Fmechanics\u002Foscillations-waves\u002Fsimple-harmonic-motion":13312,"\u002Fmechanics\u002Foscillations-waves\u002Fpendulum-motion":13313,"\u002Fmechanics\u002Foscillations-waves\u002Fdriven-oscillators":13314,"\u002Fmechanics\u002Foscillations-waves\u002Fwave-boundaries":13315,"\u002Fmechanics\u002Fthermodynamics\u002Fkinetic-theory-of-ideal-gases":13316,"\u002Fmechanics\u002Fthermodynamics\u002Ffirst-law-of-thermodynamics":13317,"\u002Fmechanics\u002Fthermodynamics\u002Fentropy-and-the-second-law":13318,"\u002Fmechanics\u002Fthermodynamics\u002Fthermal-processes":13319,"\u002Fmechanics\u002Fthermodynamics\u002Fphase-changes":13320,"\u002Fmechanics\u002Fthermodynamics\u002Fthermal-machines":13321,"\u002Fmechanics":13322,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Fcharge-and-conductors":13323,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Fcoulombs-law":13324,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Felectric-field-and-force":13325,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Felectric-field-maps":13326,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Felectric-dipoles":13327,"\u002Felectricity-and-magnetism\u002Fcontinuous-charge-distributions\u002Fcontinuous-charge-fields":13328,"\u002Felectricity-and-magnetism\u002Fcontinuous-charge-distributions\u002Fgauss-law-and-conductors":13329,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Fpoint-charge-potential":13330,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Fpotential-gradients-and-equipotentials":13331,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Felectrostatic-energy-and-pressure":13332,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Flaplace-boundary-problems":13333,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Fcontinuous-charge-potentials":13334,"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fcapacitance-fundamentals":13311,"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fcapacitor-networks":13335,"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fcapacitor-energy-and-force":13336,"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fdielectric-polarization-and-breakdown":13337,"\u002Felectricity-and-magnetism\u002Fdirect-current-circuits\u002Fcurrent-and-resistance":13307,"\u002Felectricity-and-magnetism\u002Fdirect-current-circuits\u002Fkirchhoff-network-analysis":13172,"\u002Felectricity-and-magnetism\u002Fdirect-current-circuits\u002Frc-transients":13338,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmagnetic-trajectories":13298,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fhall-effect":13339,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmagnetic-force-on-conductors":13340,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmagnetic-dipoles":13341,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmass-spectrometry":13342,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fmoving-charge-fields":13343,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fbiot-savart-law":13344,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fcircular-current-loops":13345,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Famperes-law":13346,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fgauss-law-for-magnetism":13347,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fmagnetic-materials":13272,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fmagnetic-flux":13348,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Ffaradays-law":13349,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Flenzs-law":13350,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fmotional-emf":13351,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Feddy-currents":13352,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fself-inductance":13353,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fmagnetic-energy":13354,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Frl-circuits":13355,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Fac-fundamentals":13290,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Freactance":13289,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Frlc-resonance":13356,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Fac-power":13357,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Ftransformers":13358,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Fdisplacement-current":13359,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Felectromagnetic-waves":13360,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Felectromagnetic-momentum":13361,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Fdipole-radiation":13362,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Fpolarization":13363,"\u002Felectricity-and-magnetism\u002Foptics\u002Freflection-and-refraction":13364,"\u002Felectricity-and-magnetism\u002Foptics\u002Fthin-lenses":13316,"\u002Felectricity-and-magnetism\u002Foptics\u002Fspherical-mirrors":13314,"\u002Felectricity-and-magnetism":13365,"\u002Flinear-algebra\u002Flinear-systems\u002Fsystems-and-echelon-forms":13366,"\u002Flinear-algebra\u002Flinear-systems\u002Fvector-and-matrix-equations":13367,"\u002Flinear-algebra\u002Flinear-systems\u002Fsolution-sets-and-applications":13368,"\u002Flinear-algebra\u002Flinear-systems\u002Flinear-independence":13369,"\u002Flinear-algebra\u002Flinear-systems\u002Flinear-transformations":13370,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fmatrix-operations":13371,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fmatrix-inverse-and-invertibility":13372,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fpartitioned-matrices-and-lu":13373,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fsubspaces-dimension-rank":13374,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fapplications-leontief-and-graphics":13224,"\u002Flinear-algebra\u002Fdeterminants\u002Fdeterminants-and-cofactors":13375,"\u002Flinear-algebra\u002Fdeterminants\u002Fproperties-of-determinants":13376,"\u002Flinear-algebra\u002Fdeterminants\u002Fcramer-volume-and-area":13228,"\u002Flinear-algebra\u002Fvector-spaces\u002Fvector-spaces-and-subspaces":13377,"\u002Flinear-algebra\u002Fvector-spaces\u002Fnull-and-column-spaces":13378,"\u002Flinear-algebra\u002Fvector-spaces\u002Fbases-and-independent-sets":13379,"\u002Flinear-algebra\u002Fvector-spaces\u002Fcoordinate-systems":13380,"\u002Flinear-algebra\u002Fvector-spaces\u002Fdimension-and-rank":13381,"\u002Flinear-algebra\u002Fvector-spaces\u002Fchange-of-basis":13382,"\u002Flinear-algebra\u002Fvector-spaces\u002Fdifference-equations-and-markov":13383,"\u002Flinear-algebra\u002Feigenvalues\u002Feigenvectors-and-eigenvalues":13384,"\u002Flinear-algebra\u002Feigenvalues\u002Fthe-characteristic-equation":13385,"\u002Flinear-algebra\u002Feigenvalues\u002Fdiagonalization":13386,"\u002Flinear-algebra\u002Feigenvalues\u002Feigenvectors-and-linear-transformations":13387,"\u002Flinear-algebra\u002Feigenvalues\u002Fcomplex-eigenvalues":13388,"\u002Flinear-algebra\u002Feigenvalues\u002Fdynamical-systems":13389,"\u002Flinear-algebra\u002Feigenvalues\u002Fpower-method":13390,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Finner-product-length-orthogonality":13391,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Forthogonal-sets-and-projections":13392,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Fgram-schmidt-and-qr":13393,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Fleast-squares-problems":13394,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Fleast-squares-applications":13395,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Finner-product-spaces":13396,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fdiagonalizing-symmetric-matrices":13263,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fquadratic-forms":13397,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fconstrained-optimization":13398,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fsingular-value-decomposition":13399,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fsvd-applications-pca-imaging":13400,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fnumerical-thinking-and-matrix-computation":13401,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Flu-and-cholesky":13402,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fconditioning-and-floating-point":13403,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fstability-and-error-analysis":13404,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fqr-and-numerical-least-squares":13405,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fnumerical-eigenvalues-and-svd":13406,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Faffine-combinations":13407,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Faffine-independence-and-barycentric-coordinates":13408,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Fconvex-combinations-and-convex-sets":13409,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Fhyperplanes-and-polytopes":13410,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Fcurves-and-surfaces":13411,"\u002Flinear-algebra":13412,"\u002Ftheory-of-computation":13413,"\u002Fcomputer-architecture\u002Ffoundations\u002Fbits-bytes-and-words":13414,"\u002Fcomputer-architecture\u002Ffoundations\u002Finteger-representation":13415,"\u002Fcomputer-architecture\u002Ffoundations\u002Finteger-arithmetic":13416,"\u002Fcomputer-architecture\u002Ffoundations\u002Ffloating-point":13417,"\u002Fcomputer-architecture\u002Ffoundations\u002Fboolean-algebra-and-bit-manipulation":13418,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fthe-machines-view":13419,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fdata-movement":13420,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Farithmetic-and-logic":13421,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fcontrol-flow":13422,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fprocedures":13423,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Farrays-structs-and-alignment":13424,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fmemory-layout-and-buffer-overflows":13425,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Fwhat-an-isa-is":13426,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Finstruction-formats-and-operands":13427,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Faddressing-modes":13428,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Fthe-y86-64-instruction-set":13429,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Fy86-64-programming":13430,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Ftransistors-gates-and-boolean-functions":13431,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fcombinational-logic-and-hcl":13432,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fmultiplexers-decoders-and-the-alu":13433,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fmemory-elements-latches-flip-flops-and-clocking":13434,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fregister-files-and-random-access-memory":13435,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fthe-fetch-decode-execute-cycle":13436,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fthe-seq-stages":13437,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fcontrol-logic-and-sequencing":13438,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fassembling-seq":13439,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Ftracing-a-program":13440,"\u002Fcomputer-architecture\u002Fpipelining\u002Fpipelining-principles":13441,"\u002Fcomputer-architecture\u002Fpipelining\u002Ffrom-seq-to-pipe":13442,"\u002Fcomputer-architecture\u002Fpipelining\u002Fdata-hazards-stalling-and-forwarding":13443,"\u002Fcomputer-architecture\u002Fpipelining\u002Fcontrol-hazards-and-branch-prediction":13444,"\u002Fcomputer-architecture\u002Fpipelining\u002Fthe-complete-pipe-processor":13445,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fstorage-technologies-and-the-latency-gap":13446,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Flocality":13447,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fcache-memories-direct-mapped":13448,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fset-associative-and-write-policies":13449,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fcache-performance-and-cache-friendly-code":13450,"\u002Fcomputer-architecture\u002Fvirtual-memory\u002Faddress-spaces-and-translation":13451,"\u002Fcomputer-architecture\u002Fvirtual-memory\u002Fpage-tables-and-page-faults":13452,"\u002Fcomputer-architecture\u002Fvirtual-memory\u002Fthe-tlb-and-multi-level-page-tables":13453,"\u002Fcomputer-architecture\u002Fexceptions-and-io\u002Fexceptional-control-flow":13454,"\u002Fcomputer-architecture\u002Fexceptions-and-io\u002Finterrupts-and-the-kernel":13455,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fprocesses-threads-and-parallelism":13456,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fhardware-multithreading":13457,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fcache-coherence":13458,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fmemory-consistency-and-synchronization":13459,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fmulticore-organization":13460,"\u002Fcomputer-architecture\u002Fcapstone\u002Fthe-whole-machine":13461,"\u002Fcomputer-architecture\u002Fcapstone\u002Fassembling-a-complete-cpu":13462,"\u002Fcomputer-architecture":13413,"\u002Fdifferential-equations\u002Ffoundations\u002Fmodels-and-direction-fields":13463,"\u002Fdifferential-equations\u002Ffoundations\u002Fclassification-and-terminology":13464,"\u002Fdifferential-equations\u002Ffirst-order\u002Flinear-first-order-integrating-factors":13465,"\u002Fdifferential-equations\u002Ffirst-order\u002Fseparable-and-exact":13228,"\u002Fdifferential-equations\u002Ffirst-order\u002Fmodeling-first-order":13466,"\u002Fdifferential-equations\u002Ffirst-order\u002Fautonomous-and-population-dynamics":13227,"\u002Fdifferential-equations\u002Ffirst-order\u002Fexistence-uniqueness-euler":13234,"\u002Fdifferential-equations\u002Ffirst-order\u002Ffirst-order-difference-equations":13467,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fhomogeneous-constant-coefficients":13468,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fcomplex-and-repeated-roots":13268,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fnonhomogeneous-undetermined-coefficients":13469,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fvariation-of-parameters":13470,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fmechanical-electrical-vibrations":13471,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fhigher-order-linear":13472,"\u002Fdifferential-equations\u002Fseries-solutions\u002Fpower-series-ordinary-points":13473,"\u002Fdifferential-equations\u002Fseries-solutions\u002Fregular-singular-frobenius":13474,"\u002Fdifferential-equations\u002Fseries-solutions\u002Fbessel-and-special-functions":13475,"\u002Fdifferential-equations\u002Flaplace\u002Flaplace-definition-ivps":13476,"\u002Fdifferential-equations\u002Flaplace\u002Fstep-impulse-convolution":13477,"\u002Fdifferential-equations\u002Fsystems\u002Fmatrices-eigenvalues-review":13478,"\u002Fdifferential-equations\u002Fsystems\u002Fconstant-coefficient-systems-phase-portraits":13479,"\u002Fdifferential-equations\u002Fsystems\u002Frepeated-eigenvalues-fundamental-matrices":13480,"\u002Fdifferential-equations\u002Fnumerical\u002Feuler-and-runge-kutta":13475,"\u002Fdifferential-equations\u002Fnumerical\u002Fmultistep-systems-stability":13481,"\u002Fdifferential-equations\u002Fnonlinear\u002Fphase-plane-autonomous-stability":13482,"\u002Fdifferential-equations\u002Fnonlinear\u002Flocally-linear-and-liapunov":13483,"\u002Fdifferential-equations\u002Fnonlinear\u002Fcompeting-species-predator-prey-limit-cycles":13484,"\u002Fdifferential-equations\u002Fpdes-fourier-bvp\u002Ffourier-series":13485,"\u002Fdifferential-equations\u002Fpdes-fourier-bvp\u002Fheat-wave-laplace-equations":13486,"\u002Fdifferential-equations\u002Fpdes-fourier-bvp\u002Fsturm-liouville":13487,"\u002Fdifferential-equations\u002Fhistory-variations\u002Fcalculus-of-variations":13488,"\u002Fdifferential-equations\u002Fhistory-variations\u002Fhistorical-notes":13489,"\u002Fdifferential-equations":13490,"\u002Frelativity\u002Ffoundations\u002Fspecial-relativity-postulates":13491,"\u002Frelativity\u002Ffoundations\u002Florentz-transformation-spacetime":13492,"\u002Frelativity\u002Ffoundations\u002Ftime-dilation-length-contraction":13493,"\u002Frelativity\u002Ffoundations\u002Frelativistic-momentum-energy":13494,"\u002Frelativity\u002Ffoundations\u002Fgeneral-relativity":13373,"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Fminkowski-spacetime-and-the-interval":13495,"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Ffour-vectors-and-index-notation":13496,"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Fthe-lorentz-group-and-rapidity":13497,"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Fdoppler-aberration-and-appearance":13498,"\u002Frelativity\u002Frelativistic-dynamics\u002Ffour-momentum-force-and-accelerated-motion":13499,"\u002Frelativity\u002Frelativistic-dynamics\u002Fparticle-decays-and-two-body-kinematics":13500,"\u002Frelativity\u002Frelativistic-dynamics\u002Fcollisions-thresholds-and-the-cm-frame":13243,"\u002Frelativity\u002Frelativistic-dynamics\u002Fmandelstam-variables-and-invariants":13501,"\u002Frelativity\u002Fcovariant-electrodynamics\u002Ffour-current-and-the-four-potential":13502,"\u002Frelativity\u002Fcovariant-electrodynamics\u002Fthe-electromagnetic-field-tensor":13503,"\u002Frelativity\u002Fcovariant-electrodynamics\u002Ftransformation-of-electric-and-magnetic-fields":13504,"\u002Frelativity\u002Fcovariant-electrodynamics\u002Fcovariant-maxwell-and-the-stress-energy-tensor":13505,"\u002Frelativity\u002Fcurved-spacetime\u002Fthe-equivalence-principle-formalized":13506,"\u002Frelativity\u002Fcurved-spacetime\u002Fmanifolds-vectors-and-the-metric":13507,"\u002Frelativity\u002Fcurved-spacetime\u002Fcovariant-derivative-and-christoffel-symbols":13508,"\u002Frelativity\u002Fcurved-spacetime\u002Fgeodesics-and-the-geodesic-equation":13509,"\u002Frelativity\u002Fcurved-spacetime\u002Fcurvature-riemann-and-geodesic-deviation":13510,"\u002Frelativity\u002Fcurved-spacetime\u002Fthe-einstein-field-equations":13465,"\u002Frelativity\u002Fthe-schwarzschild-solution\u002Fthe-schwarzschild-metric":13511,"\u002Frelativity\u002Fthe-schwarzschild-solution\u002Fgeodesics-and-orbits-in-schwarzschild":13512,"\u002Frelativity\u002Fthe-schwarzschild-solution\u002Flight-bending-and-null-geodesics":13513,"\u002Frelativity\u002Ftests-of-general-relativity\u002Fperihelion-precession-of-mercury":13514,"\u002Frelativity\u002Ftests-of-general-relativity\u002Fdeflection-of-light-and-gravitational-lensing":13515,"\u002Frelativity\u002Ftests-of-general-relativity\u002Fgravitational-redshift-and-shapiro-delay":13394,"\u002Frelativity\u002Ftests-of-general-relativity\u002Frelativity-in-technology-gps":13516,"\u002Frelativity\u002Fblack-holes\u002Fhorizons-and-coordinate-singularities":13517,"\u002Frelativity\u002Fblack-holes\u002Frotating-and-charged-black-holes":13404,"\u002Frelativity\u002Fblack-holes\u002Fblack-hole-thermodynamics":13518,"\u002Frelativity\u002Fgravitational-waves\u002Flinearized-gravity-and-wave-solutions":13519,"\u002Frelativity\u002Fgravitational-waves\u002Fgeneration-and-the-quadrupole-formula":13520,"\u002Frelativity\u002Fgravitational-waves\u002Fdetection-ligo-and-the-first-events":13521,"\u002Frelativity\u002Fcosmological-bridge\u002Fthe-cosmological-principle-and-flrw-metric":13522,"\u002Frelativity\u002Fcosmological-bridge\u002Ffriedmann-equations-and-cosmic-dynamics":13523,"\u002Frelativity":13524,"\u002Fphysical-computing":13413,"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fblackbody-radiation-and-the-planck-quantum":13525,"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fthe-photoelectric-effect-and-the-photon":13504,"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fx-rays-and-the-compton-effect":13526,"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fthe-old-quantum-theory-bohr-and-sommerfeld":13527,"\u002Fquantum-mechanics\u002Fmatter-waves\u002Fde-broglie-waves-and-electron-diffraction":13528,"\u002Fquantum-mechanics\u002Fmatter-waves\u002Fwave-packets-and-the-probability-interpretation":13529,"\u002Fquantum-mechanics\u002Fmatter-waves\u002Fthe-uncertainty-principle":13530,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fthe-schrodinger-equation-in-one-dimension":13531,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fthe-free-particle-and-wave-packet-dynamics":13532,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fparticle-in-infinite-and-finite-square-wells":13480,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Foperators-expectation-values-and-the-harmonic-oscillator":13404,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fthe-dirac-delta-potential":13533,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fbarrier-penetration-and-quantum-tunneling":13534,"\u002Fquantum-mechanics\u002Fformalism\u002Fhilbert-space-and-dirac-notation":13535,"\u002Fquantum-mechanics\u002Fformalism\u002Fobservables-hermitian-operators-and-eigenvalues":13536,"\u002Fquantum-mechanics\u002Fformalism\u002Fthe-postulates-and-quantum-measurement":13532,"\u002Fquantum-mechanics\u002Fformalism\u002Fposition-momentum-and-continuous-spectra":13537,"\u002Fquantum-mechanics\u002Fformalism\u002Fcommutators-and-the-generalized-uncertainty-principle":13512,"\u002Fquantum-mechanics\u002Fformalism\u002Ftime-evolution-schrodinger-and-heisenberg-pictures":13265,"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fladder-operators-and-the-number-states":13538,"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fcoherent-and-squeezed-states":13539,"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fsymmetries-generators-and-conservation-laws":13245,"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fparity-time-reversal-and-discrete-symmetries":13540,"\u002Fquantum-mechanics\u002Fangular-momentum\u002Forbital-angular-momentum-and-spherical-harmonics":13541,"\u002Fquantum-mechanics\u002Fangular-momentum\u002Fthe-angular-momentum-algebra":13542,"\u002Fquantum-mechanics\u002Fangular-momentum\u002Faddition-of-angular-momenta-and-clebsch-gordan":13543,"\u002Fquantum-mechanics\u002Fcentral-potentials\u002Fthe-schrodinger-equation-in-three-dimensions":13544,"\u002Fquantum-mechanics\u002Fcentral-potentials\u002Fthe-hydrogen-atom":13545,"\u002Fquantum-mechanics\u002Fcentral-potentials\u002Fthe-isotropic-oscillator-and-hidden-symmetry":13546,"\u002Fquantum-mechanics\u002Fspin\u002Fspin-half-pauli-matrices-and-stern-gerlach":13547,"\u002Fquantum-mechanics\u002Fspin\u002Fspin-in-a-magnetic-field-precession-and-resonance":13548,"\u002Fquantum-mechanics\u002Fspin\u002Ftwo-level-systems-and-the-bloch-sphere":13504,"\u002Fquantum-mechanics\u002Fidentical-particles\u002Fidentical-particles-and-exchange-symmetry":13549,"\u002Fquantum-mechanics\u002Fidentical-particles\u002Fthe-pauli-principle-atoms-and-the-periodic-table":13550,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Ftime-independent-perturbation-theory":13551,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Ffine-structure-and-the-real-hydrogen-atom":13538,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Fthe-zeeman-and-stark-effects":13226,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Fthe-variational-method":13552,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Fthe-wkb-approximation":13553,"\u002Fquantum-mechanics":13554,"\u002Freal-analysis\u002Ffoundations\u002Fsets-logic-functions":13488,"\u002Freal-analysis\u002Ffoundations\u002Fordered-fields-completeness":13555,"\u002Freal-analysis\u002Ffoundations\u002Fabsolute-value-bounds":13556,"\u002Freal-analysis\u002Ffoundations\u002Fintervals-uncountability":13377,"\u002Freal-analysis\u002Fsequences-series\u002Fsequences-limits":13557,"\u002Freal-analysis\u002Fsequences-series\u002Flimit-laws-monotone":13263,"\u002Freal-analysis\u002Fsequences-series\u002Flimsup-bolzano-weierstrass":13558,"\u002Freal-analysis\u002Fsequences-series\u002Fcauchy-completeness":13559,"\u002Freal-analysis\u002Fsequences-series\u002Fseries-convergence":13400,"\u002Freal-analysis\u002Fsequences-series\u002Fabsolute-conditional-rearrangement":13510,"\u002Freal-analysis\u002Fmetric-spaces\u002Fmetric-spaces-norms":13560,"\u002Freal-analysis\u002Fmetric-spaces\u002Fopen-closed-sets":13561,"\u002Freal-analysis\u002Fmetric-spaces\u002Fconvergence-completeness":13562,"\u002Freal-analysis\u002Fmetric-spaces\u002Fcompactness":13563,"\u002Freal-analysis\u002Fmetric-spaces\u002Fconnectedness":13564,"\u002Freal-analysis\u002Fcontinuity\u002Flimits-of-functions":13537,"\u002Freal-analysis\u002Fcontinuity\u002Fcontinuous-functions":13565,"\u002Freal-analysis\u002Fcontinuity\u002Fevt-ivt":13370,"\u002Freal-analysis\u002Fcontinuity\u002Funiform-continuity":13566,"\u002Freal-analysis\u002Fcontinuity\u002Fcontinuity-metric-spaces":13567,"\u002Freal-analysis\u002Fcontinuity\u002Flimits-infinity-monotone":13224,"\u002Freal-analysis\u002Fdifferentiation\u002Fthe-derivative":13568,"\u002Freal-analysis\u002Fdifferentiation\u002Fmean-value-theorem":13569,"\u002Freal-analysis\u002Fdifferentiation\u002Ftaylors-theorem":13526,"\u002Freal-analysis\u002Fdifferentiation\u002Finverse-function-1d":13253,"\u002Freal-analysis\u002Friemann-integration\u002Fdarboux-integral":13404,"\u002Freal-analysis\u002Friemann-integration\u002Fintegrability-classes":13570,"\u002Freal-analysis\u002Friemann-integration\u002Fproperties-of-the-integral":13571,"\u002Freal-analysis\u002Friemann-integration\u002Ffundamental-theorem":13389,"\u002Freal-analysis\u002Friemann-integration\u002Flog-exp-improper":13514,"\u002Freal-analysis\u002Ffunction-sequences\u002Fpointwise-uniform-convergence":13572,"\u002Freal-analysis\u002Ffunction-sequences\u002Finterchange-of-limits":13573,"\u002Freal-analysis\u002Ffunction-sequences\u002Fpower-series-weierstrass":13574,"\u002Freal-analysis\u002Ffunction-sequences\u002Fpicard-ode":13409,"\u002Freal-analysis\u002Fseveral-variables\u002Fdifferentiability-rn":13575,"\u002Freal-analysis\u002Fseveral-variables\u002Fgradient-chain-rule":13576,"\u002Freal-analysis\u002Fseveral-variables\u002Fhigher-derivatives-taylor-extrema":13577,"\u002Freal-analysis\u002Fseveral-variables\u002Finverse-implicit-theorems":13577,"\u002Freal-analysis\u002Fseveral-variables\u002Fmultiple-integrals":13578,"\u002Freal-analysis":13579,"\u002Fabstract-algebra\u002Ffoundations\u002Fsets-functions-relations":13580,"\u002Fabstract-algebra\u002Ffoundations\u002Fintegers-and-modular-arithmetic":13581,"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fgroup-axioms-and-first-examples":13582,"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fdihedral-and-symmetric-groups":13583,"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fmatrix-and-quaternion-groups":13584,"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fhomomorphisms-and-group-actions":13585,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fsubgroups-and-substructures":13586,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fcyclic-groups":13587,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fgeneration-and-subgroup-lattices":13588,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fcosets-lagrange-and-normal-subgroups":13589,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fisomorphism-theorems":13553,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fcomposition-series-and-the-alternating-group":13590,"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Factions-and-cayleys-theorem":13582,"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Fconjugation-and-the-class-equation":13499,"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Fsylow-theorems":13591,"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Fautomorphisms-and-simple-groups":13592,"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fdirect-products-and-finite-abelian-groups":13593,"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fsemidirect-products":13594,"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fnilpotent-and-solvable-groups":13595,"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fclassifying-small-groups":13596,"\u002Fabstract-algebra\u002Fring-theory\u002Frings-definitions-and-examples":13597,"\u002Fabstract-algebra\u002Fring-theory\u002Fideals-quotients-and-homomorphisms":13598,"\u002Fabstract-algebra\u002Fring-theory\u002Ffractions-and-the-chinese-remainder-theorem":13592,"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Feuclidean-domains-pids-ufds":13599,"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Fpolynomial-rings-over-fields":13568,"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Fgauss-lemma-and-unique-factorization":13600,"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Firreducibility-criteria-and-groebner":13601,"\u002Fabstract-algebra\u002Fmodule-theory\u002Fintroduction-to-modules":13602,"\u002Fabstract-algebra\u002Fmodule-theory\u002Ffree-modules-and-direct-sums":13603,"\u002Fabstract-algebra\u002Fmodule-theory\u002Ftensor-products-and-exact-sequences":13604,"\u002Fabstract-algebra\u002Fmodule-theory\u002Fvector-spaces-and-linear-maps":13605,"\u002Fabstract-algebra\u002Fmodules-over-pids\u002Fstructure-theorem-over-pids":13606,"\u002Fabstract-algebra\u002Fmodules-over-pids\u002Frational-canonical-form":13607,"\u002Fabstract-algebra\u002Fmodules-over-pids\u002Fjordan-canonical-form":13608,"\u002Fabstract-algebra\u002Ffield-theory\u002Ffield-extensions-and-algebraic-elements":13609,"\u002Fabstract-algebra\u002Ffield-theory\u002Fstraightedge-and-compass-constructions":13242,"\u002Fabstract-algebra\u002Ffield-theory\u002Fsplitting-fields-and-algebraic-closure":13610,"\u002Fabstract-algebra\u002Ffield-theory\u002Fseparable-and-cyclotomic-extensions":13611,"\u002Fabstract-algebra\u002Fgalois-theory\u002Fthe-galois-correspondence":13484,"\u002Fabstract-algebra\u002Fgalois-theory\u002Ffinite-fields":13612,"\u002Fabstract-algebra\u002Fgalois-theory\u002Fcyclotomic-and-abelian-extensions":13613,"\u002Fabstract-algebra\u002Fgalois-theory\u002Fgalois-groups-of-polynomials":13541,"\u002Fabstract-algebra\u002Fgalois-theory\u002Fsolvability-by-radicals-and-the-quintic":13613,"\u002Fabstract-algebra\u002Fcapstone\u002Fcommutative-algebra-and-algebraic-geometry":13614,"\u002Fabstract-algebra\u002Fcapstone\u002Frepresentation-and-character-theory":13615,"\u002Fabstract-algebra":13616,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fatomic-spectra-rutherford":13617,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fbohr-model-hydrogen":13618,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fx-ray-spectra-franck-hertz":13619,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fbohr-sommerfeld-old-quantum-theory":13620,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fold-quantum-theory-limits-wkb":13621,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fschrodinger-3d-hydrogen":13533,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fhydrogen-wave-functions":13622,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fradial-equation-in-full":13623,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fsymmetry-degeneracy-runge-lenz":13624,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fexpectation-values-virial":13625,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fquantum-defects-alkali-spectra":13626,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Frydberg-atoms":13627,"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Frelativistic-kinetic-correction":13628,"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Fspin-orbit-thomas-precession":13367,"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Fdarwin-term-fine-structure-formula":13496,"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Fdirac-equation-hydrogen":13242,"\u002Fatomic-physics\u002Fqed-corrections-and-hyperfine-structure\u002Flamb-shift-qed":13629,"\u002Fatomic-physics\u002Fqed-corrections-and-hyperfine-structure\u002Fhyperfine-structure-21cm":13226,"\u002Fatomic-physics\u002Fqed-corrections-and-hyperfine-structure\u002Fnuclear-effects-isotope-shift":13630,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fperiodic-table-atomic-spectra":13631,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fcentral-field-self-consistent":13264,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fidentical-particles-hartree-fock":13558,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fhelium-two-electron-atom":13632,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fls-jj-coupling-term-symbols":13633,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fhund-rules-ground-terms":13634,"\u002Fatomic-physics\u002Fatoms-in-external-fields\u002Fzeeman-effect":13635,"\u002Fatomic-physics\u002Fatoms-in-external-fields\u002Fpaschen-back-intermediate":13636,"\u002Fatomic-physics\u002Fatoms-in-external-fields\u002Fstark-effect-polarizability":13637,"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Ftime-dependent-perturbation-golden-rule":13638,"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Fdipole-approximation-einstein-coefficients":13639,"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Fselection-rules-forbidden-transitions":13640,"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Flifetimes-and-line-shapes":13641,"\u002Fatomic-physics\u002Flasers-and-spectroscopy\u002Flaser-principles":13642,"\u002Fatomic-physics\u002Flasers-and-spectroscopy\u002Fspectroscopy-techniques":13643,"\u002Fatomic-physics\u002Flasers-and-spectroscopy\u002Fline-catalog-nist-asd":13644,"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Flaser-cooling-doppler":13645,"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Fsub-doppler-trapping":13589,"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Fbose-einstein-condensation":13646,"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Foptical-clocks-precision":13647,"\u002Fatomic-physics":13648,"\u002Fdatabases":13413,"\u002Fcategory-theory\u002Ffoundations\u002Fwhat-is-a-category":13649,"\u002Fcategory-theory\u002Ffoundations\u002Fexamples-of-categories":13650,"\u002Fcategory-theory\u002Ffoundations\u002Fspecial-morphisms":13651,"\u002Fcategory-theory\u002Ffoundations\u002Ffunctors":13580,"\u002Fcategory-theory\u002Ffoundations\u002Fnatural-transformations":13652,"\u002Fcategory-theory\u002Ffoundations\u002Fsize-and-set-theory":13653,"\u002Fcategory-theory\u002Funiversal-properties\u002Funiversal-properties":13654,"\u002Fcategory-theory\u002Funiversal-properties\u002Fproducts-and-coproducts":13655,"\u002Fcategory-theory\u002Funiversal-properties\u002Fconstructions-on-categories":13656,"\u002Fcategory-theory\u002Frepresentables-yoneda\u002Frepresentable-functors":13657,"\u002Fcategory-theory\u002Frepresentables-yoneda\u002Fyoneda-lemma":13658,"\u002Fcategory-theory\u002Frepresentables-yoneda\u002Fyoneda-consequences":13659,"\u002Fcategory-theory\u002Flimits-colimits\u002Flimits":13660,"\u002Fcategory-theory\u002Flimits-colimits\u002Fproducts-equalizers-pullbacks":13661,"\u002Fcategory-theory\u002Flimits-colimits\u002Fcolimits":13662,"\u002Fcategory-theory\u002Flimits-colimits\u002Fcomputing-limits":13663,"\u002Fcategory-theory\u002Flimits-colimits\u002Flimits-and-functors":13664,"\u002Fcategory-theory\u002Fadjunctions\u002Fadjunctions":13665,"\u002Fcategory-theory\u002Fadjunctions\u002Funits-and-counits":13666,"\u002Fcategory-theory\u002Fadjunctions\u002Fadjunctions-via-universal-arrows":13667,"\u002Fcategory-theory\u002Fadjunctions\u002Ffree-forgetful-adjunctions":13668,"\u002Fcategory-theory\u002Fadjoints-limits\u002Flimits-via-adjoints":13669,"\u002Fcategory-theory\u002Fadjoints-limits\u002Fpresheaf-limits-colimits":13670,"\u002Fcategory-theory\u002Fadjoints-limits\u002Fadjoints-preserve-limits":13671,"\u002Fcategory-theory\u002Fadjoints-limits\u002Fadjoint-functor-theorem":13662,"\u002Fcategory-theory\u002Fmonads-algebras\u002Fmonads":13672,"\u002Fcategory-theory\u002Fmonads-algebras\u002Falgebras-eilenberg-moore":13673,"\u002Fcategory-theory\u002Fmonads-algebras\u002Fkleisli-and-programming":13674,"\u002Fcategory-theory\u002Fmonads-algebras\u002Falgebras-for-endofunctors":13675,"\u002Fcategory-theory\u002Fcartesian-closed-lambda\u002Fcartesian-closed-categories":13676,"\u002Fcategory-theory\u002Fcartesian-closed-lambda\u002Flambda-calculus-correspondence":13621,"\u002Fcategory-theory\u002Fcartesian-closed-lambda\u002Ffixed-points-and-recursion":13677,"\u002Fcategory-theory":13678,"\u002Fdeep-learning\u002Fmathematical-background\u002Flinear-algebra-for-deep-learning":13679,"\u002Fdeep-learning\u002Fmathematical-background\u002Fprobability-and-information-theory":13680,"\u002Fdeep-learning\u002Fmathematical-background\u002Fnumerical-computation":13681,"\u002Fdeep-learning\u002Fmathematical-background\u002Fcalculus":13682,"\u002Fdeep-learning\u002Ffoundations\u002Fwhat-is-deep-learning":13683,"\u002Fdeep-learning\u002Ffoundations\u002Fmachine-learning-refresher":13684,"\u002Fdeep-learning\u002Ffoundations\u002Flinear-models-and-the-perceptron":13643,"\u002Fdeep-learning\u002Fneural-networks\u002Fthe-multilayer-perceptron":13685,"\u002Fdeep-learning\u002Fneural-networks\u002Factivation-functions":13686,"\u002Fdeep-learning\u002Fneural-networks\u002Funiversal-approximation":13687,"\u002Fdeep-learning\u002Fneural-networks\u002Fbackpropagation":13688,"\u002Fdeep-learning\u002Fneural-networks\u002Floss-functions-and-output-units":13689,"\u002Fdeep-learning\u002Foptimization\u002Fgradient-descent-and-sgd":13690,"\u002Fdeep-learning\u002Foptimization\u002Fmomentum-and-adaptive-methods":13691,"\u002Fdeep-learning\u002Foptimization\u002Finitialization":13692,"\u002Fdeep-learning\u002Foptimization\u002Fthe-optimization-landscape":13693,"\u002Fdeep-learning\u002Foptimization\u002Fsecond-order-and-approximate-methods":13694,"\u002Fdeep-learning\u002Fregularization\u002Fregularization-overview":13695,"\u002Fdeep-learning\u002Fregularization\u002Fdropout-and-data-augmentation":13696,"\u002Fdeep-learning\u002Fregularization\u002Fearly-stopping-and-parameter-sharing":13697,"\u002Fdeep-learning\u002Fregularization\u002Fnormalization":13698,"\u002Fdeep-learning\u002Farchitectures\u002Fconvolutional-networks":13699,"\u002Fdeep-learning\u002Farchitectures\u002Fcnn-architectures":13700,"\u002Fdeep-learning\u002Farchitectures\u002Frecurrent-networks":13701,"\u002Fdeep-learning\u002Farchitectures\u002Flstm-and-gru":13702,"\u002Fdeep-learning\u002Farchitectures\u002Fattention-and-transformers":13703,"\u002Fdeep-learning\u002Farchitectures\u002Fthe-transformer-architecture":13704,"\u002Fdeep-learning\u002Farchitectures\u002Ftransformers-in-practice":13705,"\u002Fdeep-learning\u002Farchitectures\u002Fgraph-neural-networks":13706,"\u002Fdeep-learning\u002Farchitectures\u002Fstate-space-models":13707,"\u002Fdeep-learning\u002Ftheory\u002Fgeneralization-theory":13708,"\u002Fdeep-learning\u002Ftheory\u002Fadversarial-robustness":13709,"\u002Fdeep-learning\u002Ftheory\u002Fadversarial-defenses":13710,"\u002Fdeep-learning\u002Ftheory\u002Fbayesian-and-ensemble-methods":13711,"\u002Fdeep-learning\u002Ftheory\u002Fdeep-equilibrium-models":13655,"\u002Fdeep-learning\u002Fgenerative-models\u002Flinear-factor-models":13712,"\u002Fdeep-learning\u002Fgenerative-models\u002Fautoencoders":13713,"\u002Fdeep-learning\u002Fgenerative-models\u002Fvariational-autoencoders":13714,"\u002Fdeep-learning\u002Fgenerative-models\u002Fgenerative-adversarial-networks":13715,"\u002Fdeep-learning\u002Fgenerative-models\u002Fautoregressive-and-normalizing-flows":13716,"\u002Fdeep-learning\u002Fgenerative-models\u002Fenergy-based-and-boltzmann-machines":13717,"\u002Fdeep-learning\u002Fgenerative-models\u002Fdiffusion-and-score-based-models":13718,"\u002Fdeep-learning\u002Fprobabilistic-methods\u002Fstructured-probabilistic-models":13207,"\u002Fdeep-learning\u002Fprobabilistic-methods\u002Fmonte-carlo-and-mcmc":13719,"\u002Fdeep-learning\u002Fprobabilistic-methods\u002Fapproximate-inference":13720,"\u002Fdeep-learning\u002Fpractical\u002Fpractical-methodology":13443,"\u002Fdeep-learning\u002Fpractical\u002Fhyperparameters-and-debugging":13721,"\u002Fdeep-learning\u002Fpractical\u002Frepresentation-learning":13722,"\u002Fdeep-learning\u002Fpractical\u002Ftransfer-learning":13723,"\u002Fdeep-learning\u002Fpractical\u002Fapplications":13724,"\u002Fdeep-learning\u002Fpractical\u002Fmodel-compression-and-distillation":13725,"\u002Fdeep-learning\u002Fpractical\u002Fmeta-learning-and-few-shot":13726,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Flarge-language-models":13727,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fscaling-inference-and-alignment":13728,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fseq2seq-pretraining-and-bart":13729,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Ftext-to-text-transfer-and-conditional-generation":13730,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fspeech-and-audio-models":13731,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fself-supervised-speech-and-synthesis":13732,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fai-agents":13416,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fagent-memory-retrieval-and-orchestration":13733,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fmixture-of-experts":13734,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fmultimodal-models":13735,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Ffusion-and-vision-language-models":13736,"\u002Fdeep-learning\u002Freinforcement-learning\u002Ffoundations-of-reinforcement-learning":13453,"\u002Fdeep-learning\u002Freinforcement-learning\u002Fmodel-free-prediction-and-control":13737,"\u002Fdeep-learning\u002Freinforcement-learning\u002Fdeep-q-networks":13738,"\u002Fdeep-learning\u002Freinforcement-learning\u002Fpolicy-gradients-and-actor-critic":13739,"\u002Fdeep-learning\u002Freinforcement-learning\u002Frl-from-human-feedback":13740,"\u002Fdeep-learning":13413,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fequilibrium-state-variables-zeroth-law":13741,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Ffirst-law-heat-and-work":13538,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fsecond-law-entropy-and-the-carnot-bound":13742,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fthermodynamic-potentials-and-maxwell-relations":13743,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fstability-response-functions-and-the-third-law":13744,"\u002Fstatistical-mechanics\u002Ffoundations\u002Fclassical-statistics-and-equipartition":13745,"\u002Fstatistical-mechanics\u002Ffoundations\u002Fphase-space-and-liouvilles-theorem":13746,"\u002Fstatistical-mechanics\u002Ffoundations\u002Fensembles-and-the-equal-probability-postulate":13747,"\u002Fstatistical-mechanics\u002Ffoundations\u002Fstatistical-entropy-boltzmann-and-gibbs":13748,"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Fmicrocanonical-ensemble-and-entropy":13749,"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Fequilibrium-conditions-temperature-pressure-chemical-potential":13585,"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Fideal-gas-phase-space-and-the-sackur-tetrode-entropy":13750,"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Ftwo-state-systems-paramagnets-and-negative-temperature":13751,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fcanonical-ensemble-and-the-boltzmann-distribution":13752,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fpartition-function-and-the-helmholtz-free-energy":13472,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fenergy-fluctuations-and-ensemble-equivalence":13265,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fthe-einstein-solid-and-harmonic-systems":13753,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fparamagnetism-and-the-schottky-anomaly":13754,"\u002Fstatistical-mechanics\u002Fclassical-gas\u002Fideal-gas-partition-function-and-the-gibbs-paradox":13755,"\u002Fstatistical-mechanics\u002Fclassical-gas\u002Fequipartition-and-the-virial-theorem":13399,"\u002Fstatistical-mechanics\u002Fclassical-gas\u002Fmolecular-gases-rotation-and-vibration":13756,"\u002Fstatistical-mechanics\u002Fgrand-canonical\u002Fgrand-canonical-ensemble-and-the-grand-partition-function":13757,"\u002Fstatistical-mechanics\u002Fgrand-canonical\u002Fchemical-potential-fugacity-and-number-fluctuations":13477,"\u002Fstatistical-mechanics\u002Fgrand-canonical\u002Fensemble-summary-and-the-thermodynamic-web":13404,"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fquantum-statistics-bose-einstein-and-fermi-dirac":13758,"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fderiving-the-quantum-distributions":13373,"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fthe-classical-limit-and-quantum-concentration":13243,"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fideal-quantum-gases-general-framework":13259,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fbose-einstein-condensation-and-the-fermion-gas":13759,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fthe-photon-gas-and-plancks-radiation-law":13760,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fblackbody-thermodynamics-and-radiation-pressure":13761,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fphonons-and-the-debye-model":13391,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fbose-einstein-condensation-derived":13762,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fthermodynamics-of-the-bose-gas-and-superfluidity":13253,"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fthe-ideal-fermi-gas-at-zero-temperature":13763,"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fsommerfeld-expansion-and-electrons-in-metals":13764,"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fwhite-dwarfs-and-the-chandrasekhar-limit":13765,"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fneutron-stars-and-nuclear-matter":13766,"\u002Fstatistical-mechanics\u002Finteractions\u002Fthe-cluster-expansion-and-virial-coefficients":13587,"\u002Fstatistical-mechanics\u002Finteractions\u002Fthe-van-der-waals-gas-and-liquid-gas-coexistence":13767,"\u002Fstatistical-mechanics\u002Finteractions\u002Fquantum-gases-with-interactions-and-exchange":13768,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fphases-coexistence-and-classification":13769,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fthe-ising-model-and-exact-solutions":13770,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fmean-field-theory-and-the-weiss-model":13533,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fcritical-exponents-and-landau-theory":13771,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fthe-renormalization-group-idea":13511,"\u002Fstatistical-mechanics\u002Ffluctuations\u002Fthermodynamic-fluctuations-and-response":13772,"\u002Fstatistical-mechanics\u002Ffluctuations\u002Fbrownian-motion-and-the-langevin-equation":13372,"\u002Fstatistical-mechanics\u002Ffluctuations\u002Flinear-response-and-the-fluctuation-dissipation-theorem":13773,"\u002Fstatistical-mechanics":13774,"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fbonding-mechanisms":13775,"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fmolecular-orbitals-and-h2-plus":13238,"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fhydrogen-molecule-and-exchange":13498,"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fvan-der-waals-forces":13776,"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Frotational-vibrational-spectra":13777,"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Fanharmonicity-and-rovibrational-structure":13778,"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Framan-and-electronic-bands":13779,"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Flasers-and-masers":13780,"\u002Fcondensed-matter\u002Fcrystal-structure\u002Fstructure-of-solids":13781,"\u002Fcondensed-matter\u002Fcrystal-structure\u002Fbravais-lattices-and-crystal-systems":13371,"\u002Fcondensed-matter\u002Fcrystal-structure\u002Freciprocal-lattice-and-brillouin-zones":13782,"\u002Fcondensed-matter\u002Fcrystal-structure\u002Fdiffraction-and-structure-factors":13783,"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fphonon-dispersion":13784,"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fphonons-quantization-and-dos":13785,"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fdebye-einstein-heat-capacity":13514,"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fanharmonicity-and-thermal-transport":13786,"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Ffree-electron-gas-and-conduction":13787,"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Fsommerfeld-model-and-heat-capacity":13788,"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Ftransport-and-the-hall-effect":13789,"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Fscreening-and-plasmons":13790,"\u002Fcondensed-matter\u002Fband-theory\u002Fblochs-theorem-and-energy-bands":13569,"\u002Fcondensed-matter\u002Fband-theory\u002Fnearly-free-electron-model":13513,"\u002Fcondensed-matter\u002Fband-theory\u002Ftight-binding-method":13791,"\u002Fcondensed-matter\u002Fband-theory\u002Ffermi-surfaces-and-semiclassical-dynamics":13792,"\u002Fcondensed-matter\u002Fsemiconductors\u002Fsemiconductor-bands-and-junctions":13793,"\u002Fcondensed-matter\u002Fsemiconductors\u002Fintrinsic-and-extrinsic-semiconductors":13794,"\u002Fcondensed-matter\u002Fsemiconductors\u002Fcarrier-transport-and-recombination":13504,"\u002Fcondensed-matter\u002Fsemiconductors\u002Fthe-pn-junction":13795,"\u002Fcondensed-matter\u002Fsemiconductors\u002Ftransistors-and-optoelectronics":13796,"\u002Fcondensed-matter\u002Fdielectrics-and-ferroelectrics\u002Fdielectrics-and-polarization":13741,"\u002Fcondensed-matter\u002Fdielectrics-and-ferroelectrics\u002Fferroelectrics-and-piezoelectrics":13641,"\u002Fcondensed-matter\u002Fmagnetism\u002Fdiamagnetism-and-paramagnetism":13368,"\u002Fcondensed-matter\u002Fmagnetism\u002Fexchange-and-ferromagnetism":13797,"\u002Fcondensed-matter\u002Fmagnetism\u002Fantiferromagnetism-and-domains":13234,"\u002Fcondensed-matter\u002Fmagnetism\u002Fspin-waves-and-magnons":13798,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fsuperconductivity-phenomenology":13799,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Flondon-theory-and-the-meissner-effect":13379,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fginzburg-landau-theory":13800,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fbcs-theory":13634,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fjosephson-and-high-tc":13801,"\u002Fcondensed-matter\u002Fnanostructures\u002Fquantum-wells-wires-and-dots":13226,"\u002Fcondensed-matter\u002Fnanostructures\u002Finteger-quantum-hall-effect":13802,"\u002Fcondensed-matter\u002Fnanostructures\u002Ffractional-quantum-hall-and-topology":13237,"\u002Fcondensed-matter\u002Fnanostructures\u002Fgraphene-and-dirac-materials":13803,"\u002Fcondensed-matter":13554,"\u002Flogic\u002Ffoundations\u002Flogic-as-a-mathematical-model":13804,"\u002Flogic\u002Fsentential-logic\u002Fformal-languages-and-well-formed-formulas":13805,"\u002Flogic\u002Fsentential-logic\u002Ftruth-assignments-and-tautologies":13806,"\u002Flogic\u002Fsentential-logic\u002Funique-readability-and-parsing":13807,"\u002Flogic\u002Fsentential-logic\u002Finduction-and-recursion":13251,"\u002Flogic\u002Fsentential-logic\u002Fexpressive-completeness-and-normal-forms":13808,"\u002Flogic\u002Fsentential-logic\u002Fboolean-circuits":13809,"\u002Flogic\u002Fsentential-logic\u002Fcompactness-and-effectiveness":13251,"\u002Flogic\u002Ffirst-order-languages\u002Ffirst-order-languages":13810,"\u002Flogic\u002Ffirst-order-languages\u002Fstructures-truth-and-satisfaction":13664,"\u002Flogic\u002Ffirst-order-languages\u002Fdefinability-and-elementary-equivalence":13811,"\u002Flogic\u002Ffirst-order-languages\u002Fterms-substitution-and-parsing":13812,"\u002Flogic\u002Fdeductive-calculus\u002Fa-deductive-calculus":13813,"\u002Flogic\u002Fdeductive-calculus\u002Fdeduction-theorem-and-derived-rules":13811,"\u002Flogic\u002Fdeductive-calculus\u002Fsoundness":13814,"\u002Flogic\u002Fdeductive-calculus\u002Fcompleteness-and-consistency":13815,"\u002Flogic\u002Fmodels-and-theories\u002Fcompactness-and-lowenheim-skolem":13816,"\u002Flogic\u002Fmodels-and-theories\u002Ftheories-elementary-classes-and-categoricity":13817,"\u002Flogic\u002Fmodels-and-theories\u002Finterpretations-between-theories":13818,"\u002Flogic\u002Fmodels-and-theories\u002Fnonstandard-analysis":13819,"\u002Flogic\u002Farithmetic-and-definability\u002Fdefinability-in-arithmetic":13820,"\u002Flogic\u002Farithmetic-and-definability\u002Fnatural-numbers-with-successor":13821,"\u002Flogic\u002Farithmetic-and-definability\u002Fpresburger-and-reducts":13742,"\u002Flogic\u002Farithmetic-and-definability\u002Fa-subtheory-and-representability":13822,"\u002Flogic\u002Fincompleteness\u002Farithmetization-of-syntax":13815,"\u002Flogic\u002Fincompleteness\u002Fincompleteness-and-undecidability":13823,"\u002Flogic\u002Fincompleteness\u002Fsecond-incompleteness-theorem":13824,"\u002Flogic\u002Fcomputability-and-representability\u002Frecursive-functions":13457,"\u002Flogic\u002Fcomputability-and-representability\u002Frepresenting-exponentiation":13825,"\u002Flogic\u002Fsecond-order-logic\u002Fsecond-order-languages":13642,"\u002Flogic\u002Fsecond-order-logic\u002Fskolem-functions-and-many-sorted-logic":13826,"\u002Flogic\u002Fsecond-order-logic\u002Fgeneral-structures":13827,"\u002Flogic":13828,"\u002Freinforcement-learning\u002Ffoundations\u002Fwhat-is-reinforcement-learning":13829,"\u002Freinforcement-learning\u002Ffoundations\u002Fa-brief-history-of-rl":13830,"\u002Freinforcement-learning\u002Ffoundations\u002Fmulti-armed-bandits":13443,"\u002Freinforcement-learning\u002Ffoundations\u002Fbandit-exploration-algorithms":13831,"\u002Freinforcement-learning\u002Ffoundations\u002Fmarkov-decision-processes":13832,"\u002Freinforcement-learning\u002Ffoundations\u002Fvalue-functions-and-optimality":13833,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fdynamic-programming":13834,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fdp-async-and-gpi":13824,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fmonte-carlo-methods":13835,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fmonte-carlo-off-policy":13836,"\u002Freinforcement-learning\u002Ftabular-methods\u002Ftemporal-difference-learning":13837,"\u002Freinforcement-learning\u002Ftabular-methods\u002Ftd-control-sarsa-and-q-learning":13735,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fn-step-bootstrapping":13838,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fn-step-off-policy-methods":13839,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fplanning-and-learning":13840,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fplanning-focusing-and-decision-time":13841,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fdecision-time-planning":13842,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fmonte-carlo-tree-search":13460,"\u002Freinforcement-learning\u002Fapproximation\u002Fon-policy-prediction":13843,"\u002Freinforcement-learning\u002Fapproximation\u002Ffeature-construction-and-nonlinear":13844,"\u002Freinforcement-learning\u002Fapproximation\u002Fon-policy-control":13845,"\u002Freinforcement-learning\u002Fapproximation\u002Faverage-reward-control":13846,"\u002Freinforcement-learning\u002Fapproximation\u002Foff-policy-and-the-deadly-triad":13651,"\u002Freinforcement-learning\u002Fapproximation\u002Fbellman-error-and-gradient-td":13847,"\u002Freinforcement-learning\u002Fapproximation\u002Feligibility-traces":13823,"\u002Freinforcement-learning\u002Fapproximation\u002Ftrue-online-and-sarsa-lambda":13848,"\u002Freinforcement-learning\u002Fapproximation\u002Fpolicy-gradient-methods":13849,"\u002Freinforcement-learning\u002Fapproximation\u002Factor-critic-and-continuous-actions":13850,"\u002Freinforcement-learning\u002Fapproximation\u002Fleast-squares-and-memory-based-methods":13471,"\u002Freinforcement-learning\u002Fapproximation\u002Fmemory-and-kernel-methods":13851,"\u002Freinforcement-learning\u002Fapproximation\u002Foff-policy-eligibility-traces":13601,"\u002Freinforcement-learning\u002Fapproximation\u002Fstable-off-policy-traces":13852,"\u002Freinforcement-learning\u002Fdeep-rl\u002Fdeep-q-networks":13853,"\u002Freinforcement-learning\u002Fdeep-rl\u002Fdqn-improvements":13433,"\u002Freinforcement-learning\u002Fdeep-rl\u002Factor-critic-and-ppo":13854,"\u002Freinforcement-learning\u002Fdeep-rl\u002Fppo-and-continuous-control":13855,"\u002Freinforcement-learning\u002Fdeep-rl\u002Fcase-studies":13856,"\u002Freinforcement-learning\u002Fdeep-rl\u002Frl-beyond-games":13857,"\u002Freinforcement-learning\u002Fdeep-rl\u002Ffrontiers":13858,"\u002Freinforcement-learning\u002Fdeep-rl\u002Freward-design-and-open-problems":13711,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fdistributional-and-rainbow":13859,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fdistributional-and-rainbow-part-2":13860,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fcontinuous-control":13861,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fcontinuous-control-part-2":13746,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmodel-based-rl":13862,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmodel-based-rl-part-2":13863,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fexploration":13864,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fexploration-part-2":13460,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Foffline-rl":13527,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Foffline-rl-part-2":13865,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fimitation-and-inverse-rl":13866,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fimitation-and-inverse-rl-part-2":13867,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmulti-agent-rl":13868,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmulti-agent-rl-part-2":13869,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fhierarchical-rl":13870,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fhierarchical-rl-part-2":13871,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Frlhf-and-language-models":13872,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fpartial-observability-pomdps":13873,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fpartial-observability-pomdps-part-2":13874,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fsafe-and-constrained-rl":13875,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fsafe-and-constrained-rl-part-2":13876,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmeta-rl-and-generalization":13877,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fpsychology-of-reinforcement":13878,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Finstrumental-conditioning-and-control":13879,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fdopamine-and-td-error":13880,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fdopamine-in-the-brain":13881,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fanimal-learning-and-cognition":13882,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fcognitive-maps-and-planning":13883,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fneuroscience-of-reinforcement":13884,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fseveral-learning-systems":13885,"\u002Freinforcement-learning":13413,"\u002Fartificial-intelligence\u002Ffoundations\u002Fwhat-is-ai":13886,"\u002Fartificial-intelligence\u002Ffoundations\u002Ffoundations-of-ai":13887,"\u002Fartificial-intelligence\u002Ffoundations\u002Fintelligent-agents":13888,"\u002Fartificial-intelligence\u002Ffoundations\u002Fagent-architectures":13889,"\u002Fartificial-intelligence\u002Fsearch\u002Funinformed-search":13890,"\u002Fartificial-intelligence\u002Fsearch\u002Fsearch-strategies-compared":13891,"\u002Fartificial-intelligence\u002Fsearch\u002Finformed-search":13892,"\u002Fartificial-intelligence\u002Fsearch\u002Fheuristic-functions":13893,"\u002Fartificial-intelligence\u002Fsearch\u002Flocal-search":13894,"\u002Fartificial-intelligence\u002Fsearch\u002Fpopulation-and-continuous-search":13895,"\u002Fartificial-intelligence\u002Fsearch\u002Fadversarial-search":13896,"\u002Fartificial-intelligence\u002Fsearch\u002Fgames-of-chance-and-imperfect-information":13897,"\u002Fartificial-intelligence\u002Fsearch\u002Fconstraint-satisfaction":13898,"\u002Fartificial-intelligence\u002Fsearch\u002Fcsp-search-and-structure":13740,"\u002Fartificial-intelligence\u002Fsearch\u002Fsearch-under-uncertainty":13595,"\u002Fartificial-intelligence\u002Fsearch\u002Fbelief-state-and-online-search":13899,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fpropositional-logic":13900,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fpropositional-inference":13901,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Ffirst-order-logic":13902,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Ffirst-order-logic-in-use":13903,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Finference-and-resolution":13904,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Ffirst-order-resolution":13722,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fclassical-planning":13905,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fplanning-graphs-and-graphplan":13906,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fplanning-in-the-real-world":13907,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fplanning-under-uncertainty":13908,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fknowledge-representation":13909,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Freasoning-systems-and-defaults":13910,"\u002Fartificial-intelligence\u002Funcertainty\u002Fprobability-and-bayes":13911,"\u002Fartificial-intelligence\u002Funcertainty\u002Fbayes-rule-and-naive-bayes":13912,"\u002Fartificial-intelligence\u002Funcertainty\u002Fbayesian-networks":13913,"\u002Fartificial-intelligence\u002Funcertainty\u002Finference-in-bayesian-networks":13914,"\u002Fartificial-intelligence\u002Funcertainty\u002Freasoning-over-time":13915,"\u002Fartificial-intelligence\u002Funcertainty\u002Ftracking-and-data-association":13916,"\u002Fartificial-intelligence\u002Funcertainty\u002Fmaking-decisions":13654,"\u002Fartificial-intelligence\u002Funcertainty\u002Fmarkov-decision-processes":13906,"\u002Fartificial-intelligence\u002Funcertainty\u002Fdecision-networks-and-game-theory":13917,"\u002Fartificial-intelligence\u002Funcertainty\u002Fgame-theory-and-mechanism-design":13194,"\u002Fartificial-intelligence\u002Flearning\u002Flearning-from-examples":13918,"\u002Fartificial-intelligence\u002Flearning\u002Ftheory-and-model-families":13919,"\u002Fartificial-intelligence\u002Flearning\u002Fprobabilistic-learning":13920,"\u002Fartificial-intelligence\u002Flearning\u002Fexpectation-maximization":13921,"\u002Fartificial-intelligence\u002Flearning\u002Freinforcement-learning":13922,"\u002Fartificial-intelligence\u002Flearning\u002Fgeneralization-and-policy-search":13703,"\u002Fartificial-intelligence\u002Flearning\u002Fknowledge-in-learning":13923,"\u002Fartificial-intelligence\u002Flearning\u002Fknowledge-based-learning-methods":13924,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fvision-and-perception":13925,"\u002Fartificial-intelligence\u002Ffrontiers\u002Freconstructing-the-3d-world":13926,"\u002Fartificial-intelligence\u002Ffrontiers\u002Frobotics":13927,"\u002Fartificial-intelligence\u002Ffrontiers\u002Frobot-planning-and-control":13928,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fnatural-language-in-ai":13929,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fnlp-grammar-translation-and-speech":13930,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fphilosophy-and-future":13931,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fai-ethics-and-future":13932,"\u002Fartificial-intelligence":13413,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-constituents-nuclide-chart":13670,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-size-charge-distributions":13933,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-masses-binding-energy":13230,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fsemi-empirical-mass-formula":13228,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-moments-multipoles":13764,"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fnuclear-force-shell-overview":13934,"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fthe-deuteron":13256,"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fnucleon-nucleon-scattering":13566,"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fmeson-theory-isospin":13935,"\u002Fnuclear-physics\u002Fnuclear-models\u002Ffermi-gas-model":13936,"\u002Fnuclear-physics\u002Fnuclear-models\u002Fliquid-drop-collective-coordinates":13937,"\u002Fnuclear-physics\u002Fnuclear-models\u002Fshell-model-single-particle":13626,"\u002Fnuclear-physics\u002Fnuclear-models\u002Fcollective-model-rotations-vibrations":13938,"\u002Fnuclear-physics\u002Fradioactive-decay\u002Fdecay-law-modes":13939,"\u002Fnuclear-physics\u002Fradioactive-decay\u002Fdecay-kinetics-equilibrium":13940,"\u002Fnuclear-physics\u002Falpha-decay\u002Falpha-decay-gamow-theory":13825,"\u002Fnuclear-physics\u002Falpha-decay\u002Falpha-fine-structure-hindrance":13941,"\u002Fnuclear-physics\u002Fbeta-decay\u002Fbeta-decay-energetics-neutrino":13942,"\u002Fnuclear-physics\u002Fbeta-decay\u002Ffermi-theory-beta-decay":13240,"\u002Fnuclear-physics\u002Fbeta-decay\u002Fweak-interaction-parity-violation":13469,"\u002Fnuclear-physics\u002Fbeta-decay\u002Fdouble-beta-decay-neutrino-mass":13943,"\u002Fnuclear-physics\u002Fgamma-decay\u002Fgamma-multipole-radiation":13570,"\u002Fnuclear-physics\u002Fgamma-decay\u002Finternal-conversion-isomers":13944,"\u002Fnuclear-physics\u002Fgamma-decay\u002Fangular-correlations-mossbauer":13945,"\u002Fnuclear-physics\u002Fnuclear-reactions\u002Freaction-kinematics-cross-sections":13465,"\u002Fnuclear-physics\u002Fnuclear-reactions\u002Fcompound-nucleus-resonances":13553,"\u002Fnuclear-physics\u002Fnuclear-reactions\u002Fdirect-reactions-optical-model":13946,"\u002Fnuclear-physics\u002Ffission\u002Ffission-barrier-dynamics":13947,"\u002Fnuclear-physics\u002Ffission\u002Fchain-reactions-reactor-physics":13948,"\u002Fnuclear-physics\u002Ffusion-nucleosynthesis\u002Ffusion-reactions-confinement":13235,"\u002Fnuclear-physics\u002Ffusion-nucleosynthesis\u002Fstellar-nucleosynthesis":13610,"\u002Fnuclear-physics\u002Ffusion-nucleosynthesis\u002Fbig-bang-nucleosynthesis":13472,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fcharged-particle-stopping-power":13949,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fphoton-neutron-interactions":13520,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fradiation-detectors":13584,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fdosimetry-radiation-biology":13950,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fnuclear-applications-dating-medicine":13951,"\u002Fnuclear-physics":13952,"\u002Fnatural-language-processing\u002Ffoundations\u002Fwhat-is-nlp":13953,"\u002Fnatural-language-processing\u002Ffoundations\u002Fregex-and-text-normalization":13954,"\u002Fnatural-language-processing\u002Ffoundations\u002Fminimum-edit-distance":13591,"\u002Fnatural-language-processing\u002Ffoundations\u002Fn-gram-language-models":13955,"\u002Fnatural-language-processing\u002Ffoundations\u002Fsmoothing-and-backoff":13956,"\u002Fnatural-language-processing\u002Fclassification\u002Fnaive-bayes-and-sentiment":13957,"\u002Fnatural-language-processing\u002Fclassification\u002Fevaluating-classifiers":13439,"\u002Fnatural-language-processing\u002Fclassification\u002Flogistic-regression":13958,"\u002Fnatural-language-processing\u002Fclassification\u002Fsentiment-and-affect-lexicons":13959,"\u002Fnatural-language-processing\u002Fsemantics\u002Fvector-semantics-and-embeddings":13736,"\u002Fnatural-language-processing\u002Fsemantics\u002Fstatic-word-embeddings":13960,"\u002Fnatural-language-processing\u002Fsemantics\u002Fneural-language-models":13905,"\u002Fnatural-language-processing\u002Fsequences\u002Fsequence-labeling":13961,"\u002Fnatural-language-processing\u002Fsequences\u002Fcrfs-and-neural-taggers":13962,"\u002Fnatural-language-processing\u002Fsequences\u002Frnns-and-lstms":13125,"\u002Fnatural-language-processing\u002Ftransformers\u002Ftransformers-and-attention":13963,"\u002Fnatural-language-processing\u002Ftransformers\u002Fthe-transformer-architecture":13964,"\u002Fnatural-language-processing\u002Ftransformers\u002Flarge-language-models":13965,"\u002Fnatural-language-processing\u002Ftransformers\u002Fllm-pretraining-and-scaling":13966,"\u002Fnatural-language-processing\u002Ftransformers\u002Ffine-tuning-and-prompting":13417,"\u002Fnatural-language-processing\u002Ftransformers\u002Fprompting-and-alignment":13967,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fconstituency-parsing":13968,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcky-scoring-and-evaluation":13911,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fdependency-parsing":13969,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fgraph-based-and-neural-dependency-parsing":13970,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fword-senses-and-wsd":13971,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fwsd-in-practice-and-induction":13972,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fsemantic-roles-and-information-extraction":13973,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Frelations-events-and-templates":13974,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcoreference-and-discourse":13975,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcoherence-and-discourse-structure":13976,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Flogical-semantics":13835,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcompositional-semantics-and-description-logics":13977,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fsemantic-parsing":13978,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fneural-semantic-parsing":13979,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Finformation-extraction":13980,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Ftimes-events-and-templates":13981,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fdiscourse-coherence":13982,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fentity-based-and-global-coherence":13983,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fconstituency-grammars":13984,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Ftreebanks-and-lexicalized-grammars":13985,"\u002Fnatural-language-processing\u002Fapplications\u002Fmachine-translation":13986,"\u002Fnatural-language-processing\u002Fapplications\u002Fmachine-translation-decoding-and-evaluation":13987,"\u002Fnatural-language-processing\u002Fapplications\u002Fquestion-answering":13988,"\u002Fnatural-language-processing\u002Fapplications\u002Fquestion-answering-knowledge-and-llms":13689,"\u002Fnatural-language-processing\u002Fapplications\u002Fdialogue-and-chatbots":13856,"\u002Fnatural-language-processing\u002Fapplications\u002Fdialogue-systems-and-assistants":13445,"\u002Fnatural-language-processing\u002Fapplications\u002Ftext-summarization":13989,"\u002Fnatural-language-processing\u002Fapplications\u002Fabstractive-summarization-and-evaluation":13990,"\u002Fnatural-language-processing\u002Fspeech\u002Fphonetics":13991,"\u002Fnatural-language-processing\u002Fspeech\u002Facoustic-phonetics":13992,"\u002Fnatural-language-processing\u002Fspeech\u002Fautomatic-speech-recognition":13703,"\u002Fnatural-language-processing\u002Fspeech\u002Fasr-evaluation-and-applications":13993,"\u002Fnatural-language-processing":13413,"\u002Fparticle-physics\u002Ffoundations\u002Fhistorical-overview-particle-zoo":13994,"\u002Fparticle-physics\u002Ffoundations\u002Fparticle-physics-basic-concepts":13611,"\u002Fparticle-physics\u002Ffoundations\u002Ffundamental-interactions-force-carriers":13995,"\u002Fparticle-physics\u002Funits-kinematics\u002Fnatural-units-and-scales":13756,"\u002Fparticle-physics\u002Funits-kinematics\u002Ffour-vectors-invariant-mass":13996,"\u002Fparticle-physics\u002Funits-kinematics\u002Fdecay-scattering-kinematics-mandelstam":13997,"\u002Fparticle-physics\u002Funits-kinematics\u002Fcross-sections-golden-rule":13946,"\u002Fparticle-physics\u002Fsymmetries\u002Fconservation-laws-symmetries":13998,"\u002Fparticle-physics\u002Fsymmetries\u002Fdiscrete-symmetries-cpt":13999,"\u002Fparticle-physics\u002Fsymmetries\u002Fparity-violation-weak":13561,"\u002Fparticle-physics\u002Fsymmetries\u002Fsu2-su3-flavor-symmetry":13263,"\u002Fparticle-physics\u002Fquark-model\u002Feightfold-way-su3":14000,"\u002Fparticle-physics\u002Fquark-model\u002Fmeson-spectroscopy":13515,"\u002Fparticle-physics\u002Fquark-model\u002Fbaryon-spectroscopy":14001,"\u002Fparticle-physics\u002Fquark-model\u002Fcolor-confinement-exotics":13548,"\u002Fparticle-physics\u002Frelativistic-wave-equations\u002Fklein-gordon-equation":14002,"\u002Fparticle-physics\u002Frelativistic-wave-equations\u002Fdirac-equation-spinors":13802,"\u002Fparticle-physics\u002Frelativistic-wave-equations\u002Fantiparticles-hole-theory":14003,"\u002Fparticle-physics\u002Fqed\u002Ffeynman-rules-qed":14004,"\u002Fparticle-physics\u002Fqed\u002Fqed-tree-processes":13771,"\u002Fparticle-physics\u002Fqed\u002Frenormalization-running-coupling":14005,"\u002Fparticle-physics\u002Fqed\u002Felectron-g-2":13234,"\u002Fparticle-physics\u002Fweak-interaction\u002Fva-structure-weak":14006,"\u002Fparticle-physics\u002Fweak-interaction\u002Fw-z-bosons-decays":14007,"\u002Fparticle-physics\u002Fweak-interaction\u002Fckm-matrix":14008,"\u002Fparticle-physics\u002Fweak-interaction\u002Fcp-violation-kaons-b-mesons":13552,"\u002Fparticle-physics\u002Fqcd\u002Fcolor-su3-gluons":13768,"\u002Fparticle-physics\u002Fqcd\u002Fasymptotic-freedom-confinement":14009,"\u002Fparticle-physics\u002Fqcd\u002Fdeep-inelastic-scattering-partons":14010,"\u002Fparticle-physics\u002Fqcd\u002Fjets-hadronization":13627,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Felectroweak-su2-u1":14011,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fspontaneous-symmetry-breaking":13672,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fhiggs-mechanism":14012,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fhiggs-boson-discovery":14013,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fstandard-model":13561,"\u002Fparticle-physics\u002Fneutrinos\u002Fneutrino-oscillations":13594,"\u002Fparticle-physics\u002Fneutrinos\u002Fneutrino-mass-pmns":14014,"\u002Fparticle-physics\u002Fneutrinos\u002Fdirac-majorana-experiments":14015,"\u002Fparticle-physics\u002Fexperiment\u002Faccelerators-luminosity":14016,"\u002Fparticle-physics\u002Fexperiment\u002Fdetectors-subsystems":13650,"\u002Fparticle-physics\u002Fexperiment\u002Fhow-discoveries-are-made":14017,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fbeyond-standard-model":13391,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fgrand-unified-theories":14018,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fsupersymmetry":13494,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fhierarchy-problem-naturalness":14019,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fdark-matter-candidates":14020,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fmatter-antimatter-open-questions":13818,"\u002Fparticle-physics":14021,"\u002Fastrophysics-cosmology\u002Forientation\u002Fthe-sun-and-stars":13612,"\u002Fastrophysics-cosmology\u002Forientation\u002Fstellar-death-final-states":13766,"\u002Fastrophysics-cosmology\u002Forientation\u002Fgalaxies-and-cosmology":14022,"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Fmagnitudes-fluxes-and-the-distance-modulus":13551,"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Fstellar-spectra-and-spectral-classification":14023,"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Ftelescopes-and-detectors-across-the-spectrum":13611,"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Fthe-cosmic-distance-ladder":13523,"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fblackbody-radiation-and-specific-intensity":14024,"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fradiative-transfer-and-the-transfer-equation":14025,"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fspectral-line-formation-and-broadening":14026,"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fopacity-and-the-rosseland-mean":14027,"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fhydrostatic-equilibrium-and-the-virial-theorem":14028,"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fthe-equations-of-stellar-structure":13817,"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fthe-equation-of-state-and-polytropes":13748,"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fthe-standard-solar-model":14029,"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fthermonuclear-reaction-rates-and-the-gamow-peak":14030,"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fhydrogen-burning-pp-chains-and-cno":14031,"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fhelium-burning-and-the-triple-alpha-process":14032,"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fadvanced-burning-and-neutron-capture-nucleosynthesis":13945,"\u002Fastrophysics-cosmology\u002Fism-and-star-formation\u002Fphases-of-the-interstellar-medium":14033,"\u002Fastrophysics-cosmology\u002Fism-and-star-formation\u002Fmolecular-clouds-and-gravitational-collapse":13507,"\u002Fastrophysics-cosmology\u002Fism-and-star-formation\u002Fprotostars-and-the-pre-main-sequence":13570,"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fthe-main-sequence-and-its-structure":13552,"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fpost-main-sequence-low-mass-evolution":14034,"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fthe-evolution-of-massive-stars":13636,"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fstellar-pulsation-and-the-instability-strip":13252,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fwhite-dwarfs-and-the-chandrasekhar-limit":14035,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fcore-collapse-supernovae":14036,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fthermonuclear-supernovae-type-ia":13767,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fneutron-stars-and-pulsars":14037,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fblack-holes-schwarzschild-and-kerr":14038,"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Fbinary-systems-and-mass-transfer":14039,"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Faccreting-compact-objects":14040,"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Fgravitational-waves-from-inspiraling-binaries":13233,"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Fmultimessenger-astronomy-and-gamma-ray-bursts":14041,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fthe-milky-way":14042,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fgalaxy-morphology-and-classification":13494,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fgalaxy-rotation-curves-and-dark-matter":14043,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Factive-galactic-nuclei-and-supermassive-black-holes":14044,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fgalaxy-clusters-and-large-scale-structure":13760,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fthe-expanding-universe-and-hubbles-law":13415,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fthe-frw-metric-and-cosmological-redshift":14045,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fthe-friedmann-equations-and-cosmic-dynamics":14046,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fcosmological-models-and-distances":13672,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fdark-energy-and-the-accelerating-universe":13658,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fthe-thermal-history-of-the-universe":13670,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fbig-bang-nucleosynthesis":13767,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Frecombination-and-the-cosmic-microwave-background":14047,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fcmb-anisotropies-and-cosmological-parameters":13232,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fcosmic-inflation":13424,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fstructure-formation-and-the-growth-of-perturbations":14048,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fdark-matter-dark-energy-and-open-questions":13511,"\u002Fastrophysics-cosmology":13616,"\u002Fcolophon":14049,"\u002F":13413},4250,4808,3626,2682,4109,4786,3878,3875,3751,3415,4067,3153,3000,4042,5461,5808,3961,3749,4327,5067,4246,4655,4154,5436,2640,4003,3601,2158,4331,4189,2273,3252,4633,4964,4172,3131,5524,3160,4031,2309,4207,3226,2648,4842,5340,3307,5701,4977,4039,2615,3472,4460,3848,4075,4400,3382,3010,3602,3737,3740,3707,3922,5191,4043,3804,4542,4214,5062,2850,4361,3443,3627,4044,3766,4140,3860,4006,5199,4334,5234,3651,5509,5680,153,1375,1073,1093,1125,1146,1014,1132,876,1541,1189,1173,984,1402,1301,950,1268,1063,1107,1408,1161,925,1012,866,964,1090,1142,1085,1020,1207,973,980,728,764,1225,1329,796,929,801,878,774,1044,1488,1175,1130,890,814,870,154,4073,5140,4961,5127,4870,5382,5195,4955,5369,4501,5576,3824,4132,4289,4307,4570,3403,5084,5105,5201,5116,5341,5175,5368,5188,5211,5499,5155,4981,5125,5415,5255,5304,5130,5167,5552,5164,5094,5239,5036,5190,5004,5099,5035,5159,5088,5026,4937,5023,5264,5244,133,5114,5078,5043,5312,5170,5342,5139,5151,5049,5212,5013,5068,5079,5102,5121,5081,5029,5379,5854,5110,2139,3798,5055,5364,4984,4935,4895,4972,5289,5112,5156,4987,5031,5025,5149,5302,5042,5002,4979,4922,4960,5279,126,1877,1180,1129,907,958,1112,1300,1053,1250,1181,1241,1234,966,1050,734,1190,484,1082,926,733,761,571,607,798,804,952,977,731,784,645,771,1017,742,1004,1000,1562,1254,1288,1101,1011,1486,1061,856,992,1169,988,137,0,2037,1782,2384,2254,2123,2332,1643,1714,2089,1751,1367,1660,2511,1998,1892,1854,1791,2438,2487,1917,2375,2525,2266,1845,2275,1810,1631,2310,2166,2233,2113,2505,2347,2672,2112,2473,2592,2380,3013,2513,3256,3218,2194,2173,2205,2326,2081,3342,3152,1799,1670,1027,960,1095,1291,986,897,1209,1055,1817,1801,1593,1465,1196,1464,1201,1230,1435,1684,1461,1926,1500,1409,1284,1774,1869,162,1487,1122,1188,1001,1351,982,1005,979,1325,1046,943,1279,824,1008,989,1798,1277,1025,987,1043,1211,1074,981,939,1002,739,1139,1108,1013,1070,978,1458,1317,157,1357,1077,2355,1116,1037,1178,1637,1314,1109,1056,1702,1474,1071,1158,832,993,1404,1024,1068,1339,1106,1264,1248,913,1848,1328,1633,1224,1143,135,1378,959,1028,998,911,1527,1203,1266,1483,1165,990,938,965,1257,1418,1099,942,1352,956,1035,1398,1003,1094,1292,138,1721,1827,1449,1354,1148,1184,1285,1281,1213,1290,1271,1252,1274,1778,1591,1503,1437,1571,1584,1957,1117,1781,1648,1342,1667,1510,1965,1607,1365,1849,1259,1303,1356,1238,2208,1564,173,1671,1286,1227,1638,1529,668,1078,918,709,865,880,940,1534,1015,874,922,841,794,1194,822,1105,1658,1359,1296,1438,1921,1844,1570,1429,1324,1400,140,1787,1558,1654,1492,1747,2224,2002,2009,1323,1349,1785,1573,1722,1829,1353,1548,1552,1583,1624,1585,1245,1364,1514,1343,1397,1355,2211,1481,1770,160,2388,2293,2256,2552,2569,2478,2039,2496,2578,2814,2519,2461,2587,2492,2714,3278,2654,3050,2447,2849,2238,2369,2061,2214,2602,2563,2186,2985,2749,3364,2038,2282,2409,2126,2573,2206,2176,2268,2182,2402,2705,2633,2414,2213,2801,3313,3410,3195,1952,2017,1509,2537,2645,2027,2415,2838,2356,1906,3184,2950,2807,2954,1683,1316,1034,1138,1763,1822,1705,1246,1701,1097,1104,1187,1032,1083,1228,916,1489,1033,1652,997,692,837,1023,888,864,1089,1231,1214,1675,1156,1075,1520,1309,139,1205,1051,735,1123,1072,915,567,768,825,1253,983,1007,762,1058,861,862,971,1208,1149,1145,1029,1084,927,810,838,857,807,936,949,2321,1622,1069,1113,1057,854,1958,1528,1618,2049,1432,1679,1796,1685,1346,1275,1476,1505,1610,2018,1599,1215,1838,1909,132,3902,2215,2240,3266,3208,3073,2454,2969,2451,1875,2728,1884,2371,2516,2842,1690,1904,2346,3146,1386,2607,1966,2668,1665,2885,1606,2577,3074,2869,2403,2433,2082,1939,1587,2460,2747,2032,2642,1619,3123,1993,2090,2339,3829,1737,2622,2340,2322,3828,4409,2305,3411,2510,4527,3030,3569,3043,2457,1946,2277,2044,2909,1693,1945,2093,2399,2115,2898,2742,2242,3895,3378,3376,2769,2223,3062,3262,2651,2949,2768,3128,2423,1977,2087,2866,3388,2830,2210,2489,2884,3945,2099,2713,3402,1692,2931,4195,3989,3206,4391,3004,3704,3494,2902,999,881,901,919,748,869,1018,1045,1049,1333,954,1092,1019,976,1771,1480,1396,953,1026,161,3533,2495,1818,3007,2595,3427,3537,2216,1895,2304,1739,2073,1962,2203,1767,2666,2264,2276,2852,1807,3735,1560,4144,1669,1676,1972,2418,3291,1525,2040,2766,2337,2220,2800,3001,2078,1759,2836,1896,2026,1758,1543,1047,896,946,1060,1384,1482,815,1414,1322,1440,1240,1468,1098,1133,847,1009,1381,1052,1191,1258,1370,1712,1441,1199,957,1079,150,1262,1417,1368,1219,1136,1064,1463,1636,1059,931,1115,1736,1174,1376,1363,1411,1247,1746,1313,1299,1617,1102,1076,1495,1265,1193,1263,80,{"\u002Falgorithms\u002Ffoundations\u002Fwhat-is-an-algorithm":14051,"\u002Falgorithms\u002Ffoundations\u002Fproof-techniques":14056,"\u002Falgorithms\u002Ffoundations\u002Fasymptotic-analysis":14060,"\u002Falgorithms\u002Ffoundations\u002Fgrowth-rates-and-loop-analysis":14064,"\u002Falgorithms\u002Ffoundations\u002Frecurrences":14068,"\u002Falgorithms\u002Ffoundations\u002Famortized-analysis":14072,"\u002Falgorithms\u002Fdivide-and-conquer\u002Fmergesort":14076,"\u002Falgorithms\u002Fdivide-and-conquer\u002Fquicksort":14081,"\u002Falgorithms\u002Fdivide-and-conquer\u002Fselection":14085,"\u002Falgorithms\u002Fdivide-and-conquer\u002Ffast-multiplication":14089,"\u002Falgorithms\u002Fsorting\u002Fheaps-and-heapsort":14093,"\u002Falgorithms\u002Fsorting\u002Fsorting-lower-bounds":14098,"\u002Falgorithms\u002Fsorting\u002Flinear-time-sorting":14102,"\u002Falgorithms\u002Fsorting\u002Fexternal-sorting":14106,"\u002Falgorithms\u002Fdata-structures\u002Felementary-structures":14110,"\u002Falgorithms\u002Fdata-structures\u002Fhash-tables":14115,"\u002Falgorithms\u002Fdata-structures\u002Fbinary-search-trees":14119,"\u002Falgorithms\u002Fdata-structures\u002Favl-trees":14123,"\u002Falgorithms\u002Fdata-structures\u002Fbalanced-trees":14127,"\u002Falgorithms\u002Fdata-structures\u002Funion-find":14131,"\u002Falgorithms\u002Fdata-structures\u002Ffenwick-and-segment-trees":14135,"\u002Falgorithms\u002Fdata-structures\u002Fspatial-data-structures":14139,"\u002Falgorithms\u002Fdata-structures\u002Fskip-lists-and-probabilistic-structures":14143,"\u002Falgorithms\u002Fdata-structures\u002Fb-trees":14147,"\u002Falgorithms\u002Fdata-structures\u002Fdata-stream-algorithms":14151,"\u002Falgorithms\u002Fdata-structures\u002Fstreaming-sketches":14155,"\u002Falgorithms\u002Fsequences\u002Ftwo-pointers-and-windows":14159,"\u002Falgorithms\u002Fsequences\u002Fprefix-sums":14164,"\u002Falgorithms\u002Fsequences\u002Fmonotonic-stacks":14168,"\u002Falgorithms\u002Fsequences\u002Fbinary-search-on-the-answer":14172,"\u002Falgorithms\u002Fsequences\u002Fstring-matching":14176,"\u002Falgorithms\u002Fsequences\u002Fkmp-and-z-function":14180,"\u002Falgorithms\u002Fsequences\u002Ftries":14184,"\u002Falgorithms\u002Fsequences\u002Fsuffix-arrays-and-aho-corasick":14188,"\u002Falgorithms\u002Fgraphs\u002Frepresentations-and-traversal":14192,"\u002Falgorithms\u002Fgraphs\u002Fdepth-first-search":14197,"\u002Falgorithms\u002Fgraphs\u002Ftopological-sort-and-scc":14201,"\u002Falgorithms\u002Fgraphs\u002Fminimum-spanning-trees":14205,"\u002Falgorithms\u002Fgraphs\u002Fkruskal-and-prim":14209,"\u002Falgorithms\u002Fgraphs\u002Fshortest-paths":14213,"\u002Falgorithms\u002Fgraphs\u002Fall-pairs-and-negative-weights":14217,"\u002Falgorithms\u002Fgraphs\u002Fnetwork-flow":14221,"\u002Falgorithms\u002Fgraphs\u002Fmax-flow-min-cut":14225,"\u002Falgorithms\u002Fgraphs\u002Fbridges-and-articulation-points":14229,"\u002Falgorithms\u002Fgraphs\u002Flowest-common-ancestor":14233,"\u002Falgorithms\u002Fgraphs\u002Ftwo-sat":14237,"\u002Falgorithms\u002Fgraphs\u002Feulerian-tours":14241,"\u002Falgorithms\u002Fgraphs\u002Fbipartite-matching":14245,"\u002Falgorithms\u002Fgreedy\u002Fthe-greedy-method":14249,"\u002Falgorithms\u002Fgreedy\u002Fscheduling-and-intervals":14254,"\u002Falgorithms\u002Fgreedy\u002Fhuffman-codes":14258,"\u002Falgorithms\u002Fgreedy\u002Fmatroids":14262,"\u002Falgorithms\u002Fgreedy\u002Fstable-matching":14266,"\u002Falgorithms\u002Fdynamic-programming\u002Fprinciples":14270,"\u002Falgorithms\u002Fdynamic-programming\u002Fsequence-dp":14275,"\u002Falgorithms\u002Fdynamic-programming\u002Flongest-increasing-subsequence":14279,"\u002Falgorithms\u002Fdynamic-programming\u002Fknapsack":14283,"\u002Falgorithms\u002Fdynamic-programming\u002Fcoin-change-and-unbounded":14287,"\u002Falgorithms\u002Fdynamic-programming\u002Finterval-dp":14291,"\u002Falgorithms\u002Fdynamic-programming\u002Ftree-dp":14295,"\u002Falgorithms\u002Fdynamic-programming\u002Fbitmask-dp":14299,"\u002Falgorithms\u002Fdynamic-programming\u002Fdp-optimizations":14303,"\u002Falgorithms\u002Fdynamic-programming\u002Fdp-on-graphs":14307,"\u002Falgorithms\u002Fdynamic-programming\u002Fdigit-and-probability-dp":14311,"\u002Falgorithms\u002Fbacktracking\u002Fbacktracking-fundamentals":14315,"\u002Falgorithms\u002Fbacktracking\u002Fconstraint-search":14320,"\u002Falgorithms\u002Fbacktracking\u002Fbranch-and-bound":14324,"\u002Falgorithms\u002Fbacktracking\u002Fgraph-backtracking":14328,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fnumber-theory-basics":14332,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fmodular-exponentiation-and-primality":14337,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fsieve-and-factorization":14341,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fcombinatorics":14345,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fmatrix-exponentiation":14349,"\u002Falgorithms\u002Fmathematical-algorithms\u002Ffast-fourier-transform":14353,"\u002Falgorithms\u002Fmathematical-algorithms\u002Fgradient-descent":14357,"\u002Falgorithms\u002Fcomputational-geometry\u002Fgeometric-primitives":14361,"\u002Falgorithms\u002Fcomputational-geometry\u002Fconvex-hull":14366,"\u002Falgorithms\u002Fcomputational-geometry\u002Fsweep-line":14370,"\u002Falgorithms\u002Fcomputational-geometry\u002Fpolygons-and-proximity":14374,"\u002Falgorithms\u002Fintractability\u002Fp-np-reductions":14378,"\u002Falgorithms\u002Fintractability\u002Fnp-completeness":14383,"\u002Falgorithms\u002Fintractability\u002Fcoping-with-hardness":14387,"\u002Falgorithms\u002Fintractability\u002Fapproximation-algorithms":14391,"\u002Falgorithms":14395,"\u002Fcalculus\u002Flimits-and-continuity\u002Ffunctions-and-models":14398,"\u002Fcalculus\u002Flimits-and-continuity\u002Fthe-limit-of-a-function":14403,"\u002Fcalculus\u002Flimits-and-continuity\u002Flimit-laws-and-the-precise-definition":14407,"\u002Fcalculus\u002Flimits-and-continuity\u002Fcontinuity":14411,"\u002Fcalculus\u002Fderivatives\u002Fthe-derivative-and-rates-of-change":14415,"\u002Fcalculus\u002Fderivatives\u002Fdifferentiation-rules-and-the-chain-rule":14420,"\u002Fcalculus\u002Fderivatives\u002Fimplicit-differentiation-and-related-rates":14424,"\u002Fcalculus\u002Fderivatives\u002Flinear-approximations-and-differentials":14428,"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fextrema-and-the-mean-value-theorem":14432,"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fhow-derivatives-shape-a-graph":14437,"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fcurve-sketching-and-optimization":14441,"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fnewtons-method-and-antiderivatives":14445,"\u002Fcalculus\u002Fintegrals\u002Farea-and-the-definite-integral":14449,"\u002Fcalculus\u002Fintegrals\u002Fthe-fundamental-theorem-of-calculus":14454,"\u002Fcalculus\u002Fintegrals\u002Fthe-substitution-rule":14458,"\u002Fcalculus\u002Fapplications-of-integration\u002Fareas-and-volumes":14462,"\u002Fcalculus\u002Fapplications-of-integration\u002Fwork-average-value-and-arc-length":14467,"\u002Fcalculus\u002Fapplications-of-integration\u002Fphysics-economics-and-probability":14471,"\u002Fcalculus\u002Fexponential-logarithmic-and-inverse-functions\u002Finverse-functions-logarithms-and-exponentials":14475,"\u002Fcalculus\u002Fexponential-logarithmic-and-inverse-functions\u002Fgrowth-decay-inverse-trig-and-hyperbolic-functions":14480,"\u002Fcalculus\u002Fexponential-logarithmic-and-inverse-functions\u002Flhospitals-rule":14484,"\u002Fcalculus\u002Ftechniques-of-integration\u002Fintegration-by-parts":14488,"\u002Fcalculus\u002Ftechniques-of-integration\u002Ftrigonometric-integrals-and-substitution":14493,"\u002Fcalculus\u002Ftechniques-of-integration\u002Fpartial-fractions-and-integration-strategy":14497,"\u002Fcalculus\u002Ftechniques-of-integration\u002Fapproximate-and-improper-integrals":14501,"\u002Fcalculus\u002Fparametric-and-polar\u002Fparametric-curves-and-their-calculus":14505,"\u002Fcalculus\u002Fparametric-and-polar\u002Fpolar-coordinates":14510,"\u002Fcalculus\u002Fparametric-and-polar\u002Fconic-sections":14514,"\u002Fcalculus\u002Fsequences-and-series\u002Fsequences":14518,"\u002Fcalculus\u002Fsequences-and-series\u002Fseries-and-the-integral-test":14522,"\u002Fcalculus\u002Fsequences-and-series\u002Fthe-convergence-tests":14526,"\u002Fcalculus\u002Fsequences-and-series\u002Fpower-series":14530,"\u002Fcalculus\u002Fsequences-and-series\u002Ftaylor-and-maclaurin-series":14534,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fvectors-and-the-dot-product":14538,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fthe-cross-product-lines-and-planes":14543,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fcylinders-and-quadric-surfaces":14547,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fvector-functions-and-space-curves":14551,"\u002Fcalculus\u002Fvectors-and-space-curves\u002Farc-length-curvature-and-motion":14555,"\u002Fcalculus\u002Fpartial-derivatives\u002Ffunctions-of-several-variables":14559,"\u002Fcalculus\u002Fpartial-derivatives\u002Fpartial-derivatives":14564,"\u002Fcalculus\u002Fpartial-derivatives\u002Ftangent-planes-and-the-chain-rule":14567,"\u002Fcalculus\u002Fpartial-derivatives\u002Fdirectional-derivatives-and-the-gradient":14571,"\u002Fcalculus\u002Fpartial-derivatives\u002Foptimization-and-lagrange-multipliers":14575,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fdouble-integrals":14579,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Ftriple-integrals-and-coordinate-systems":14584,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fvector-fields-and-line-integrals":14588,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fgreens-theorem-curl-and-divergence":14592,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fsurface-integrals":14596,"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fstokes-and-the-divergence-theorem":14600,"\u002Fcalculus":14604,"\u002Fmechanics\u002Ffoundations\u002Fmeasurement-and-dimensions":14607,"\u002Fmechanics\u002Ffoundations\u002Fvector-algebra":14611,"\u002Fmechanics\u002Fkinematics\u002Fone-dimensional-motion":14615,"\u002Fmechanics\u002Fkinematics\u002Fmotion-graphs":14620,"\u002Fmechanics\u002Fkinematics\u002Fprojectile-motion":14624,"\u002Fmechanics\u002Fkinematics\u002Frelative-motion":14628,"\u002Fmechanics\u002Fkinematics\u002Fcircular-motion":14632,"\u002Fmechanics\u002Fdynamics\u002Fnewtons-laws":14636,"\u002Fmechanics\u002Fdynamics\u002Ffree-body-diagrams":14641,"\u002Fmechanics\u002Fdynamics\u002Ffriction-and-curved-motion":14645,"\u002Fmechanics\u002Fdynamics\u002Fnumerical-dynamics":14649,"\u002Fmechanics\u002Fdynamics\u002Fcenter-of-mass-systems":14653,"\u002Fmechanics\u002Fenergy\u002Fwork-and-kinetic-energy":14657,"\u002Fmechanics\u002Fenergy\u002Fpotential-energy":14662,"\u002Fmechanics\u002Fenergy\u002Fmultiparticle-work":14666,"\u002Fmechanics\u002Fenergy\u002Fmass-energy-and-binding":14670,"\u002Fmechanics\u002Fenergy\u002Fphotons-and-quantization":14674,"\u002Fmechanics\u002Fmomentum\u002Fmomentum-and-collisions":14678,"\u002Fmechanics\u002Fmomentum\u002Fcenter-of-mass-collisions":14683,"\u002Fmechanics\u002Fmomentum\u002Frocket-propulsion":14687,"\u002Fmechanics\u002Frotation\u002Frotational-inertia":14691,"\u002Fmechanics\u002Frotation\u002Frotational-dynamics":14696,"\u002Fmechanics\u002Frotation\u002Frolling-motion":14700,"\u002Fmechanics\u002Frotation\u002Fangular-momentum":14704,"\u002Fmechanics\u002Frotation\u002Frolling-resistance":14708,"\u002Fmechanics\u002Frotation\u002Fgyroscopic-precession":14712,"\u002Fmechanics\u002Fgravity-and-matter\u002Fkeplerian-orbits":14716,"\u002Fmechanics\u002Fgravity-and-matter\u002Fgravitational-fields":14721,"\u002Fmechanics\u002Fgravity-and-matter\u002Fstatic-equilibrium":14725,"\u002Fmechanics\u002Fgravity-and-matter\u002Ffluid-statics":14729,"\u002Fmechanics\u002Fgravity-and-matter\u002Ffluid-flow":14733,"\u002Fmechanics\u002Fgravity-and-matter\u002Forbital-motion":14737,"\u002Fmechanics\u002Fgravity-and-matter\u002Fstress-and-elasticity":14741,"\u002Fmechanics\u002Foscillations-waves\u002Fdamped-oscillators":14745,"\u002Fmechanics\u002Foscillations-waves\u002Ftravelling-waves":14750,"\u002Fmechanics\u002Foscillations-waves\u002Fwave-superposition":14754,"\u002Fmechanics\u002Foscillations-waves\u002Fstanding-waves":14758,"\u002Fmechanics\u002Foscillations-waves\u002Fsound-waves":14762,"\u002Fmechanics\u002Foscillations-waves\u002Fdoppler-effect":14766,"\u002Fmechanics\u002Foscillations-waves\u002Fwave-packets":14770,"\u002Fmechanics\u002Foscillations-waves\u002Fbeats-and-coupling":14774,"\u002Fmechanics\u002Foscillations-waves\u002Fsimple-harmonic-motion":14778,"\u002Fmechanics\u002Foscillations-waves\u002Fpendulum-motion":14782,"\u002Fmechanics\u002Foscillations-waves\u002Fdriven-oscillators":14786,"\u002Fmechanics\u002Foscillations-waves\u002Fwave-boundaries":14790,"\u002Fmechanics\u002Fthermodynamics\u002Fkinetic-theory-of-ideal-gases":14794,"\u002Fmechanics\u002Fthermodynamics\u002Ffirst-law-of-thermodynamics":14799,"\u002Fmechanics\u002Fthermodynamics\u002Fentropy-and-the-second-law":14803,"\u002Fmechanics\u002Fthermodynamics\u002Fthermal-processes":14807,"\u002Fmechanics\u002Fthermodynamics\u002Fphase-changes":14811,"\u002Fmechanics\u002Fthermodynamics\u002Fthermal-machines":14815,"\u002Fmechanics":14819,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Fcharge-and-conductors":14822,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Fcoulombs-law":14827,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Felectric-field-and-force":14831,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Felectric-field-maps":14835,"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Felectric-dipoles":14839,"\u002Felectricity-and-magnetism\u002Fcontinuous-charge-distributions\u002Fcontinuous-charge-fields":14843,"\u002Felectricity-and-magnetism\u002Fcontinuous-charge-distributions\u002Fgauss-law-and-conductors":14848,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Fpoint-charge-potential":14852,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Fpotential-gradients-and-equipotentials":14857,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Felectrostatic-energy-and-pressure":14861,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Flaplace-boundary-problems":14865,"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Fcontinuous-charge-potentials":14869,"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fcapacitance-fundamentals":14873,"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fcapacitor-networks":14878,"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fcapacitor-energy-and-force":14882,"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fdielectric-polarization-and-breakdown":14886,"\u002Felectricity-and-magnetism\u002Fdirect-current-circuits\u002Fcurrent-and-resistance":14890,"\u002Felectricity-and-magnetism\u002Fdirect-current-circuits\u002Fkirchhoff-network-analysis":14895,"\u002Felectricity-and-magnetism\u002Fdirect-current-circuits\u002Frc-transients":14899,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmagnetic-trajectories":14903,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fhall-effect":14908,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmagnetic-force-on-conductors":14912,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmagnetic-dipoles":14916,"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmass-spectrometry":14920,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fmoving-charge-fields":14924,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fbiot-savart-law":14929,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fcircular-current-loops":14933,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Famperes-law":14937,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fgauss-law-for-magnetism":14941,"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fmagnetic-materials":14945,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fmagnetic-flux":14949,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Ffaradays-law":14954,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Flenzs-law":14958,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fmotional-emf":14962,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Feddy-currents":14966,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fself-inductance":14970,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fmagnetic-energy":14974,"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Frl-circuits":14978,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Fac-fundamentals":14982,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Freactance":14987,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Frlc-resonance":14991,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Fac-power":14995,"\u002Felectricity-and-magnetism\u002Falternating-current\u002Ftransformers":14999,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Fdisplacement-current":15003,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Felectromagnetic-waves":15008,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Felectromagnetic-momentum":15012,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Fdipole-radiation":15016,"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Fpolarization":15020,"\u002Felectricity-and-magnetism\u002Foptics\u002Freflection-and-refraction":15024,"\u002Felectricity-and-magnetism\u002Foptics\u002Fthin-lenses":15029,"\u002Felectricity-and-magnetism\u002Foptics\u002Fspherical-mirrors":15033,"\u002Felectricity-and-magnetism":15037,"\u002Flinear-algebra\u002Flinear-systems\u002Fsystems-and-echelon-forms":15040,"\u002Flinear-algebra\u002Flinear-systems\u002Fvector-and-matrix-equations":15045,"\u002Flinear-algebra\u002Flinear-systems\u002Fsolution-sets-and-applications":15049,"\u002Flinear-algebra\u002Flinear-systems\u002Flinear-independence":15053,"\u002Flinear-algebra\u002Flinear-systems\u002Flinear-transformations":15057,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fmatrix-operations":15061,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fmatrix-inverse-and-invertibility":15066,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fpartitioned-matrices-and-lu":15070,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fsubspaces-dimension-rank":15074,"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fapplications-leontief-and-graphics":15078,"\u002Flinear-algebra\u002Fdeterminants\u002Fdeterminants-and-cofactors":15082,"\u002Flinear-algebra\u002Fdeterminants\u002Fproperties-of-determinants":15087,"\u002Flinear-algebra\u002Fdeterminants\u002Fcramer-volume-and-area":15091,"\u002Flinear-algebra\u002Fvector-spaces\u002Fvector-spaces-and-subspaces":15095,"\u002Flinear-algebra\u002Fvector-spaces\u002Fnull-and-column-spaces":15100,"\u002Flinear-algebra\u002Fvector-spaces\u002Fbases-and-independent-sets":15104,"\u002Flinear-algebra\u002Fvector-spaces\u002Fcoordinate-systems":15108,"\u002Flinear-algebra\u002Fvector-spaces\u002Fdimension-and-rank":15112,"\u002Flinear-algebra\u002Fvector-spaces\u002Fchange-of-basis":15116,"\u002Flinear-algebra\u002Fvector-spaces\u002Fdifference-equations-and-markov":15120,"\u002Flinear-algebra\u002Feigenvalues\u002Feigenvectors-and-eigenvalues":15124,"\u002Flinear-algebra\u002Feigenvalues\u002Fthe-characteristic-equation":15129,"\u002Flinear-algebra\u002Feigenvalues\u002Fdiagonalization":15133,"\u002Flinear-algebra\u002Feigenvalues\u002Feigenvectors-and-linear-transformations":15137,"\u002Flinear-algebra\u002Feigenvalues\u002Fcomplex-eigenvalues":15141,"\u002Flinear-algebra\u002Feigenvalues\u002Fdynamical-systems":15145,"\u002Flinear-algebra\u002Feigenvalues\u002Fpower-method":15149,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Finner-product-length-orthogonality":15153,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Forthogonal-sets-and-projections":15158,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Fgram-schmidt-and-qr":15162,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Fleast-squares-problems":15166,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Fleast-squares-applications":15170,"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Finner-product-spaces":15174,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fdiagonalizing-symmetric-matrices":15178,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fquadratic-forms":15183,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fconstrained-optimization":15187,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fsingular-value-decomposition":15191,"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fsvd-applications-pca-imaging":15195,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fnumerical-thinking-and-matrix-computation":15199,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Flu-and-cholesky":15204,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fconditioning-and-floating-point":15208,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fstability-and-error-analysis":15212,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fqr-and-numerical-least-squares":15216,"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fnumerical-eigenvalues-and-svd":15220,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Faffine-combinations":15224,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Faffine-independence-and-barycentric-coordinates":15229,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Fconvex-combinations-and-convex-sets":15233,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Fhyperplanes-and-polytopes":15237,"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Fcurves-and-surfaces":15241,"\u002Flinear-algebra":15245,"\u002Ftheory-of-computation":15248,"\u002Fcomputer-architecture\u002Ffoundations\u002Fbits-bytes-and-words":15251,"\u002Fcomputer-architecture\u002Ffoundations\u002Finteger-representation":15255,"\u002Fcomputer-architecture\u002Ffoundations\u002Finteger-arithmetic":15259,"\u002Fcomputer-architecture\u002Ffoundations\u002Ffloating-point":15263,"\u002Fcomputer-architecture\u002Ffoundations\u002Fboolean-algebra-and-bit-manipulation":15267,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fthe-machines-view":15271,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fdata-movement":15276,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Farithmetic-and-logic":15280,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fcontrol-flow":15284,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fprocedures":15288,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Farrays-structs-and-alignment":15292,"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fmemory-layout-and-buffer-overflows":15296,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Fwhat-an-isa-is":15300,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Finstruction-formats-and-operands":15305,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Faddressing-modes":15309,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Fthe-y86-64-instruction-set":15313,"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Fy86-64-programming":15317,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Ftransistors-gates-and-boolean-functions":15321,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fcombinational-logic-and-hcl":15326,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fmultiplexers-decoders-and-the-alu":15330,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fmemory-elements-latches-flip-flops-and-clocking":15334,"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fregister-files-and-random-access-memory":15338,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fthe-fetch-decode-execute-cycle":15342,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fthe-seq-stages":15347,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fcontrol-logic-and-sequencing":15351,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fassembling-seq":15355,"\u002Fcomputer-architecture\u002Fprocessor-design\u002Ftracing-a-program":15359,"\u002Fcomputer-architecture\u002Fpipelining\u002Fpipelining-principles":15363,"\u002Fcomputer-architecture\u002Fpipelining\u002Ffrom-seq-to-pipe":15368,"\u002Fcomputer-architecture\u002Fpipelining\u002Fdata-hazards-stalling-and-forwarding":15372,"\u002Fcomputer-architecture\u002Fpipelining\u002Fcontrol-hazards-and-branch-prediction":15376,"\u002Fcomputer-architecture\u002Fpipelining\u002Fthe-complete-pipe-processor":15380,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fstorage-technologies-and-the-latency-gap":15384,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Flocality":15389,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fcache-memories-direct-mapped":15393,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fset-associative-and-write-policies":15397,"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fcache-performance-and-cache-friendly-code":15401,"\u002Fcomputer-architecture\u002Fvirtual-memory\u002Faddress-spaces-and-translation":15405,"\u002Fcomputer-architecture\u002Fvirtual-memory\u002Fpage-tables-and-page-faults":15410,"\u002Fcomputer-architecture\u002Fvirtual-memory\u002Fthe-tlb-and-multi-level-page-tables":15414,"\u002Fcomputer-architecture\u002Fexceptions-and-io\u002Fexceptional-control-flow":15418,"\u002Fcomputer-architecture\u002Fexceptions-and-io\u002Finterrupts-and-the-kernel":15423,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fprocesses-threads-and-parallelism":15427,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fhardware-multithreading":15432,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fcache-coherence":15436,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fmemory-consistency-and-synchronization":15440,"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fmulticore-organization":15444,"\u002Fcomputer-architecture\u002Fcapstone\u002Fthe-whole-machine":15448,"\u002Fcomputer-architecture\u002Fcapstone\u002Fassembling-a-complete-cpu":15453,"\u002Fcomputer-architecture":15457,"\u002Fdifferential-equations\u002Ffoundations\u002Fmodels-and-direction-fields":15460,"\u002Fdifferential-equations\u002Ffoundations\u002Fclassification-and-terminology":15464,"\u002Fdifferential-equations\u002Ffirst-order\u002Flinear-first-order-integrating-factors":15468,"\u002Fdifferential-equations\u002Ffirst-order\u002Fseparable-and-exact":15473,"\u002Fdifferential-equations\u002Ffirst-order\u002Fmodeling-first-order":15477,"\u002Fdifferential-equations\u002Ffirst-order\u002Fautonomous-and-population-dynamics":15481,"\u002Fdifferential-equations\u002Ffirst-order\u002Fexistence-uniqueness-euler":15485,"\u002Fdifferential-equations\u002Ffirst-order\u002Ffirst-order-difference-equations":15489,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fhomogeneous-constant-coefficients":15493,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fcomplex-and-repeated-roots":15498,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fnonhomogeneous-undetermined-coefficients":15502,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fvariation-of-parameters":15506,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fmechanical-electrical-vibrations":15510,"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fhigher-order-linear":15514,"\u002Fdifferential-equations\u002Fseries-solutions\u002Fpower-series-ordinary-points":15518,"\u002Fdifferential-equations\u002Fseries-solutions\u002Fregular-singular-frobenius":15523,"\u002Fdifferential-equations\u002Fseries-solutions\u002Fbessel-and-special-functions":15527,"\u002Fdifferential-equations\u002Flaplace\u002Flaplace-definition-ivps":15531,"\u002Fdifferential-equations\u002Flaplace\u002Fstep-impulse-convolution":15536,"\u002Fdifferential-equations\u002Fsystems\u002Fmatrices-eigenvalues-review":15540,"\u002Fdifferential-equations\u002Fsystems\u002Fconstant-coefficient-systems-phase-portraits":15545,"\u002Fdifferential-equations\u002Fsystems\u002Frepeated-eigenvalues-fundamental-matrices":15549,"\u002Fdifferential-equations\u002Fnumerical\u002Feuler-and-runge-kutta":15553,"\u002Fdifferential-equations\u002Fnumerical\u002Fmultistep-systems-stability":15558,"\u002Fdifferential-equations\u002Fnonlinear\u002Fphase-plane-autonomous-stability":15562,"\u002Fdifferential-equations\u002Fnonlinear\u002Flocally-linear-and-liapunov":15567,"\u002Fdifferential-equations\u002Fnonlinear\u002Fcompeting-species-predator-prey-limit-cycles":15571,"\u002Fdifferential-equations\u002Fpdes-fourier-bvp\u002Ffourier-series":15575,"\u002Fdifferential-equations\u002Fpdes-fourier-bvp\u002Fheat-wave-laplace-equations":15580,"\u002Fdifferential-equations\u002Fpdes-fourier-bvp\u002Fsturm-liouville":15584,"\u002Fdifferential-equations\u002Fhistory-variations\u002Fcalculus-of-variations":15588,"\u002Fdifferential-equations\u002Fhistory-variations\u002Fhistorical-notes":15593,"\u002Fdifferential-equations":15597,"\u002Frelativity\u002Ffoundations\u002Fspecial-relativity-postulates":15600,"\u002Frelativity\u002Ffoundations\u002Florentz-transformation-spacetime":15605,"\u002Frelativity\u002Ffoundations\u002Ftime-dilation-length-contraction":15609,"\u002Frelativity\u002Ffoundations\u002Frelativistic-momentum-energy":15613,"\u002Frelativity\u002Ffoundations\u002Fgeneral-relativity":15617,"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Fminkowski-spacetime-and-the-interval":15621,"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Ffour-vectors-and-index-notation":15626,"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Fthe-lorentz-group-and-rapidity":15630,"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Fdoppler-aberration-and-appearance":15634,"\u002Frelativity\u002Frelativistic-dynamics\u002Ffour-momentum-force-and-accelerated-motion":15638,"\u002Frelativity\u002Frelativistic-dynamics\u002Fparticle-decays-and-two-body-kinematics":15643,"\u002Frelativity\u002Frelativistic-dynamics\u002Fcollisions-thresholds-and-the-cm-frame":15647,"\u002Frelativity\u002Frelativistic-dynamics\u002Fmandelstam-variables-and-invariants":15651,"\u002Frelativity\u002Fcovariant-electrodynamics\u002Ffour-current-and-the-four-potential":15655,"\u002Frelativity\u002Fcovariant-electrodynamics\u002Fthe-electromagnetic-field-tensor":15660,"\u002Frelativity\u002Fcovariant-electrodynamics\u002Ftransformation-of-electric-and-magnetic-fields":15664,"\u002Frelativity\u002Fcovariant-electrodynamics\u002Fcovariant-maxwell-and-the-stress-energy-tensor":15668,"\u002Frelativity\u002Fcurved-spacetime\u002Fthe-equivalence-principle-formalized":15672,"\u002Frelativity\u002Fcurved-spacetime\u002Fmanifolds-vectors-and-the-metric":15677,"\u002Frelativity\u002Fcurved-spacetime\u002Fcovariant-derivative-and-christoffel-symbols":15681,"\u002Frelativity\u002Fcurved-spacetime\u002Fgeodesics-and-the-geodesic-equation":15685,"\u002Frelativity\u002Fcurved-spacetime\u002Fcurvature-riemann-and-geodesic-deviation":15689,"\u002Frelativity\u002Fcurved-spacetime\u002Fthe-einstein-field-equations":15693,"\u002Frelativity\u002Fthe-schwarzschild-solution\u002Fthe-schwarzschild-metric":15697,"\u002Frelativity\u002Fthe-schwarzschild-solution\u002Fgeodesics-and-orbits-in-schwarzschild":15702,"\u002Frelativity\u002Fthe-schwarzschild-solution\u002Flight-bending-and-null-geodesics":15706,"\u002Frelativity\u002Ftests-of-general-relativity\u002Fperihelion-precession-of-mercury":15710,"\u002Frelativity\u002Ftests-of-general-relativity\u002Fdeflection-of-light-and-gravitational-lensing":15715,"\u002Frelativity\u002Ftests-of-general-relativity\u002Fgravitational-redshift-and-shapiro-delay":15719,"\u002Frelativity\u002Ftests-of-general-relativity\u002Frelativity-in-technology-gps":15723,"\u002Frelativity\u002Fblack-holes\u002Fhorizons-and-coordinate-singularities":15727,"\u002Frelativity\u002Fblack-holes\u002Frotating-and-charged-black-holes":15732,"\u002Frelativity\u002Fblack-holes\u002Fblack-hole-thermodynamics":15736,"\u002Frelativity\u002Fgravitational-waves\u002Flinearized-gravity-and-wave-solutions":15740,"\u002Frelativity\u002Fgravitational-waves\u002Fgeneration-and-the-quadrupole-formula":15745,"\u002Frelativity\u002Fgravitational-waves\u002Fdetection-ligo-and-the-first-events":15749,"\u002Frelativity\u002Fcosmological-bridge\u002Fthe-cosmological-principle-and-flrw-metric":15753,"\u002Frelativity\u002Fcosmological-bridge\u002Ffriedmann-equations-and-cosmic-dynamics":15758,"\u002Frelativity":15762,"\u002Fphysical-computing":15765,"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fblackbody-radiation-and-the-planck-quantum":15768,"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fthe-photoelectric-effect-and-the-photon":15773,"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fx-rays-and-the-compton-effect":15777,"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fthe-old-quantum-theory-bohr-and-sommerfeld":15781,"\u002Fquantum-mechanics\u002Fmatter-waves\u002Fde-broglie-waves-and-electron-diffraction":15785,"\u002Fquantum-mechanics\u002Fmatter-waves\u002Fwave-packets-and-the-probability-interpretation":15790,"\u002Fquantum-mechanics\u002Fmatter-waves\u002Fthe-uncertainty-principle":15794,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fthe-schrodinger-equation-in-one-dimension":15798,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fthe-free-particle-and-wave-packet-dynamics":15803,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fparticle-in-infinite-and-finite-square-wells":15807,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Foperators-expectation-values-and-the-harmonic-oscillator":15811,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fthe-dirac-delta-potential":15815,"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fbarrier-penetration-and-quantum-tunneling":15819,"\u002Fquantum-mechanics\u002Fformalism\u002Fhilbert-space-and-dirac-notation":15823,"\u002Fquantum-mechanics\u002Fformalism\u002Fobservables-hermitian-operators-and-eigenvalues":15828,"\u002Fquantum-mechanics\u002Fformalism\u002Fthe-postulates-and-quantum-measurement":15832,"\u002Fquantum-mechanics\u002Fformalism\u002Fposition-momentum-and-continuous-spectra":15836,"\u002Fquantum-mechanics\u002Fformalism\u002Fcommutators-and-the-generalized-uncertainty-principle":15840,"\u002Fquantum-mechanics\u002Fformalism\u002Ftime-evolution-schrodinger-and-heisenberg-pictures":15844,"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fladder-operators-and-the-number-states":15848,"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fcoherent-and-squeezed-states":15853,"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fsymmetries-generators-and-conservation-laws":15857,"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fparity-time-reversal-and-discrete-symmetries":15861,"\u002Fquantum-mechanics\u002Fangular-momentum\u002Forbital-angular-momentum-and-spherical-harmonics":15865,"\u002Fquantum-mechanics\u002Fangular-momentum\u002Fthe-angular-momentum-algebra":15869,"\u002Fquantum-mechanics\u002Fangular-momentum\u002Faddition-of-angular-momenta-and-clebsch-gordan":15873,"\u002Fquantum-mechanics\u002Fcentral-potentials\u002Fthe-schrodinger-equation-in-three-dimensions":15877,"\u002Fquantum-mechanics\u002Fcentral-potentials\u002Fthe-hydrogen-atom":15882,"\u002Fquantum-mechanics\u002Fcentral-potentials\u002Fthe-isotropic-oscillator-and-hidden-symmetry":15886,"\u002Fquantum-mechanics\u002Fspin\u002Fspin-half-pauli-matrices-and-stern-gerlach":15890,"\u002Fquantum-mechanics\u002Fspin\u002Fspin-in-a-magnetic-field-precession-and-resonance":15895,"\u002Fquantum-mechanics\u002Fspin\u002Ftwo-level-systems-and-the-bloch-sphere":15899,"\u002Fquantum-mechanics\u002Fidentical-particles\u002Fidentical-particles-and-exchange-symmetry":15903,"\u002Fquantum-mechanics\u002Fidentical-particles\u002Fthe-pauli-principle-atoms-and-the-periodic-table":15908,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Ftime-independent-perturbation-theory":15912,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Ffine-structure-and-the-real-hydrogen-atom":15917,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Fthe-zeeman-and-stark-effects":15921,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Fthe-variational-method":15925,"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Fthe-wkb-approximation":15929,"\u002Fquantum-mechanics":15933,"\u002Freal-analysis\u002Ffoundations\u002Fsets-logic-functions":15936,"\u002Freal-analysis\u002Ffoundations\u002Fordered-fields-completeness":15941,"\u002Freal-analysis\u002Ffoundations\u002Fabsolute-value-bounds":15945,"\u002Freal-analysis\u002Ffoundations\u002Fintervals-uncountability":15949,"\u002Freal-analysis\u002Fsequences-series\u002Fsequences-limits":15953,"\u002Freal-analysis\u002Fsequences-series\u002Flimit-laws-monotone":15958,"\u002Freal-analysis\u002Fsequences-series\u002Flimsup-bolzano-weierstrass":15962,"\u002Freal-analysis\u002Fsequences-series\u002Fcauchy-completeness":15966,"\u002Freal-analysis\u002Fsequences-series\u002Fseries-convergence":15970,"\u002Freal-analysis\u002Fsequences-series\u002Fabsolute-conditional-rearrangement":15974,"\u002Freal-analysis\u002Fmetric-spaces\u002Fmetric-spaces-norms":15978,"\u002Freal-analysis\u002Fmetric-spaces\u002Fopen-closed-sets":15983,"\u002Freal-analysis\u002Fmetric-spaces\u002Fconvergence-completeness":15987,"\u002Freal-analysis\u002Fmetric-spaces\u002Fcompactness":15991,"\u002Freal-analysis\u002Fmetric-spaces\u002Fconnectedness":15995,"\u002Freal-analysis\u002Fcontinuity\u002Flimits-of-functions":15999,"\u002Freal-analysis\u002Fcontinuity\u002Fcontinuous-functions":16003,"\u002Freal-analysis\u002Fcontinuity\u002Fevt-ivt":16007,"\u002Freal-analysis\u002Fcontinuity\u002Funiform-continuity":16011,"\u002Freal-analysis\u002Fcontinuity\u002Fcontinuity-metric-spaces":16015,"\u002Freal-analysis\u002Fcontinuity\u002Flimits-infinity-monotone":16019,"\u002Freal-analysis\u002Fdifferentiation\u002Fthe-derivative":16023,"\u002Freal-analysis\u002Fdifferentiation\u002Fmean-value-theorem":16028,"\u002Freal-analysis\u002Fdifferentiation\u002Ftaylors-theorem":16032,"\u002Freal-analysis\u002Fdifferentiation\u002Finverse-function-1d":16036,"\u002Freal-analysis\u002Friemann-integration\u002Fdarboux-integral":16040,"\u002Freal-analysis\u002Friemann-integration\u002Fintegrability-classes":16045,"\u002Freal-analysis\u002Friemann-integration\u002Fproperties-of-the-integral":16049,"\u002Freal-analysis\u002Friemann-integration\u002Ffundamental-theorem":16053,"\u002Freal-analysis\u002Friemann-integration\u002Flog-exp-improper":16056,"\u002Freal-analysis\u002Ffunction-sequences\u002Fpointwise-uniform-convergence":16060,"\u002Freal-analysis\u002Ffunction-sequences\u002Finterchange-of-limits":16065,"\u002Freal-analysis\u002Ffunction-sequences\u002Fpower-series-weierstrass":16069,"\u002Freal-analysis\u002Ffunction-sequences\u002Fpicard-ode":16073,"\u002Freal-analysis\u002Fseveral-variables\u002Fdifferentiability-rn":16077,"\u002Freal-analysis\u002Fseveral-variables\u002Fgradient-chain-rule":16082,"\u002Freal-analysis\u002Fseveral-variables\u002Fhigher-derivatives-taylor-extrema":16086,"\u002Freal-analysis\u002Fseveral-variables\u002Finverse-implicit-theorems":16090,"\u002Freal-analysis\u002Fseveral-variables\u002Fmultiple-integrals":16094,"\u002Freal-analysis":16098,"\u002Fabstract-algebra\u002Ffoundations\u002Fsets-functions-relations":16101,"\u002Fabstract-algebra\u002Ffoundations\u002Fintegers-and-modular-arithmetic":16105,"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fgroup-axioms-and-first-examples":16109,"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fdihedral-and-symmetric-groups":16114,"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fmatrix-and-quaternion-groups":16118,"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fhomomorphisms-and-group-actions":16122,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fsubgroups-and-substructures":16126,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fcyclic-groups":16131,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fgeneration-and-subgroup-lattices":16135,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fcosets-lagrange-and-normal-subgroups":16139,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fisomorphism-theorems":16143,"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fcomposition-series-and-the-alternating-group":16147,"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Factions-and-cayleys-theorem":16151,"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Fconjugation-and-the-class-equation":16156,"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Fsylow-theorems":16160,"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Fautomorphisms-and-simple-groups":16164,"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fdirect-products-and-finite-abelian-groups":16168,"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fsemidirect-products":16173,"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fnilpotent-and-solvable-groups":16177,"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fclassifying-small-groups":16181,"\u002Fabstract-algebra\u002Fring-theory\u002Frings-definitions-and-examples":16185,"\u002Fabstract-algebra\u002Fring-theory\u002Fideals-quotients-and-homomorphisms":16190,"\u002Fabstract-algebra\u002Fring-theory\u002Ffractions-and-the-chinese-remainder-theorem":16194,"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Feuclidean-domains-pids-ufds":16198,"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Fpolynomial-rings-over-fields":16203,"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Fgauss-lemma-and-unique-factorization":16207,"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Firreducibility-criteria-and-groebner":16211,"\u002Fabstract-algebra\u002Fmodule-theory\u002Fintroduction-to-modules":16215,"\u002Fabstract-algebra\u002Fmodule-theory\u002Ffree-modules-and-direct-sums":16220,"\u002Fabstract-algebra\u002Fmodule-theory\u002Ftensor-products-and-exact-sequences":16224,"\u002Fabstract-algebra\u002Fmodule-theory\u002Fvector-spaces-and-linear-maps":16228,"\u002Fabstract-algebra\u002Fmodules-over-pids\u002Fstructure-theorem-over-pids":16232,"\u002Fabstract-algebra\u002Fmodules-over-pids\u002Frational-canonical-form":16237,"\u002Fabstract-algebra\u002Fmodules-over-pids\u002Fjordan-canonical-form":16241,"\u002Fabstract-algebra\u002Ffield-theory\u002Ffield-extensions-and-algebraic-elements":16245,"\u002Fabstract-algebra\u002Ffield-theory\u002Fstraightedge-and-compass-constructions":16250,"\u002Fabstract-algebra\u002Ffield-theory\u002Fsplitting-fields-and-algebraic-closure":16254,"\u002Fabstract-algebra\u002Ffield-theory\u002Fseparable-and-cyclotomic-extensions":16258,"\u002Fabstract-algebra\u002Fgalois-theory\u002Fthe-galois-correspondence":16262,"\u002Fabstract-algebra\u002Fgalois-theory\u002Ffinite-fields":16267,"\u002Fabstract-algebra\u002Fgalois-theory\u002Fcyclotomic-and-abelian-extensions":16271,"\u002Fabstract-algebra\u002Fgalois-theory\u002Fgalois-groups-of-polynomials":16275,"\u002Fabstract-algebra\u002Fgalois-theory\u002Fsolvability-by-radicals-and-the-quintic":16279,"\u002Fabstract-algebra\u002Fcapstone\u002Fcommutative-algebra-and-algebraic-geometry":16283,"\u002Fabstract-algebra\u002Fcapstone\u002Frepresentation-and-character-theory":16288,"\u002Fabstract-algebra":16292,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fatomic-spectra-rutherford":16295,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fbohr-model-hydrogen":16300,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fx-ray-spectra-franck-hertz":16304,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fbohr-sommerfeld-old-quantum-theory":16308,"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fold-quantum-theory-limits-wkb":16312,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fschrodinger-3d-hydrogen":16316,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fhydrogen-wave-functions":16321,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fradial-equation-in-full":16325,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fsymmetry-degeneracy-runge-lenz":16329,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fexpectation-values-virial":16333,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fquantum-defects-alkali-spectra":16337,"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Frydberg-atoms":16341,"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Frelativistic-kinetic-correction":16345,"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Fspin-orbit-thomas-precession":16350,"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Fdarwin-term-fine-structure-formula":16354,"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Fdirac-equation-hydrogen":16358,"\u002Fatomic-physics\u002Fqed-corrections-and-hyperfine-structure\u002Flamb-shift-qed":16362,"\u002Fatomic-physics\u002Fqed-corrections-and-hyperfine-structure\u002Fhyperfine-structure-21cm":16367,"\u002Fatomic-physics\u002Fqed-corrections-and-hyperfine-structure\u002Fnuclear-effects-isotope-shift":16371,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fperiodic-table-atomic-spectra":16375,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fcentral-field-self-consistent":16380,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fidentical-particles-hartree-fock":16384,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fhelium-two-electron-atom":16388,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fls-jj-coupling-term-symbols":16392,"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fhund-rules-ground-terms":16396,"\u002Fatomic-physics\u002Fatoms-in-external-fields\u002Fzeeman-effect":16400,"\u002Fatomic-physics\u002Fatoms-in-external-fields\u002Fpaschen-back-intermediate":16405,"\u002Fatomic-physics\u002Fatoms-in-external-fields\u002Fstark-effect-polarizability":16409,"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Ftime-dependent-perturbation-golden-rule":16413,"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Fdipole-approximation-einstein-coefficients":16418,"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Fselection-rules-forbidden-transitions":16422,"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Flifetimes-and-line-shapes":16426,"\u002Fatomic-physics\u002Flasers-and-spectroscopy\u002Flaser-principles":16430,"\u002Fatomic-physics\u002Flasers-and-spectroscopy\u002Fspectroscopy-techniques":16435,"\u002Fatomic-physics\u002Flasers-and-spectroscopy\u002Fline-catalog-nist-asd":16439,"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Flaser-cooling-doppler":16443,"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Fsub-doppler-trapping":16448,"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Fbose-einstein-condensation":16452,"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Foptical-clocks-precision":16456,"\u002Fatomic-physics":16460,"\u002Fdatabases":16463,"\u002Fcategory-theory\u002Ffoundations\u002Fwhat-is-a-category":16466,"\u002Fcategory-theory\u002Ffoundations\u002Fexamples-of-categories":16470,"\u002Fcategory-theory\u002Ffoundations\u002Fspecial-morphisms":16474,"\u002Fcategory-theory\u002Ffoundations\u002Ffunctors":16478,"\u002Fcategory-theory\u002Ffoundations\u002Fnatural-transformations":16482,"\u002Fcategory-theory\u002Ffoundations\u002Fsize-and-set-theory":16486,"\u002Fcategory-theory\u002Funiversal-properties\u002Funiversal-properties":16490,"\u002Fcategory-theory\u002Funiversal-properties\u002Fproducts-and-coproducts":16495,"\u002Fcategory-theory\u002Funiversal-properties\u002Fconstructions-on-categories":16499,"\u002Fcategory-theory\u002Frepresentables-yoneda\u002Frepresentable-functors":16503,"\u002Fcategory-theory\u002Frepresentables-yoneda\u002Fyoneda-lemma":16508,"\u002Fcategory-theory\u002Frepresentables-yoneda\u002Fyoneda-consequences":16512,"\u002Fcategory-theory\u002Flimits-colimits\u002Flimits":16516,"\u002Fcategory-theory\u002Flimits-colimits\u002Fproducts-equalizers-pullbacks":16521,"\u002Fcategory-theory\u002Flimits-colimits\u002Fcolimits":16525,"\u002Fcategory-theory\u002Flimits-colimits\u002Fcomputing-limits":16529,"\u002Fcategory-theory\u002Flimits-colimits\u002Flimits-and-functors":16533,"\u002Fcategory-theory\u002Fadjunctions\u002Fadjunctions":16537,"\u002Fcategory-theory\u002Fadjunctions\u002Funits-and-counits":16542,"\u002Fcategory-theory\u002Fadjunctions\u002Fadjunctions-via-universal-arrows":16546,"\u002Fcategory-theory\u002Fadjunctions\u002Ffree-forgetful-adjunctions":16550,"\u002Fcategory-theory\u002Fadjoints-limits\u002Flimits-via-adjoints":16554,"\u002Fcategory-theory\u002Fadjoints-limits\u002Fpresheaf-limits-colimits":16559,"\u002Fcategory-theory\u002Fadjoints-limits\u002Fadjoints-preserve-limits":16563,"\u002Fcategory-theory\u002Fadjoints-limits\u002Fadjoint-functor-theorem":16567,"\u002Fcategory-theory\u002Fmonads-algebras\u002Fmonads":16571,"\u002Fcategory-theory\u002Fmonads-algebras\u002Falgebras-eilenberg-moore":16576,"\u002Fcategory-theory\u002Fmonads-algebras\u002Fkleisli-and-programming":16580,"\u002Fcategory-theory\u002Fmonads-algebras\u002Falgebras-for-endofunctors":16584,"\u002Fcategory-theory\u002Fcartesian-closed-lambda\u002Fcartesian-closed-categories":16588,"\u002Fcategory-theory\u002Fcartesian-closed-lambda\u002Flambda-calculus-correspondence":16593,"\u002Fcategory-theory\u002Fcartesian-closed-lambda\u002Ffixed-points-and-recursion":16597,"\u002Fcategory-theory":16601,"\u002Fdeep-learning\u002Fmathematical-background\u002Flinear-algebra-for-deep-learning":16604,"\u002Fdeep-learning\u002Fmathematical-background\u002Fprobability-and-information-theory":16608,"\u002Fdeep-learning\u002Fmathematical-background\u002Fnumerical-computation":16612,"\u002Fdeep-learning\u002Fmathematical-background\u002Fcalculus":16616,"\u002Fdeep-learning\u002Ffoundations\u002Fwhat-is-deep-learning":16619,"\u002Fdeep-learning\u002Ffoundations\u002Fmachine-learning-refresher":16623,"\u002Fdeep-learning\u002Ffoundations\u002Flinear-models-and-the-perceptron":16627,"\u002Fdeep-learning\u002Fneural-networks\u002Fthe-multilayer-perceptron":16631,"\u002Fdeep-learning\u002Fneural-networks\u002Factivation-functions":16636,"\u002Fdeep-learning\u002Fneural-networks\u002Funiversal-approximation":16640,"\u002Fdeep-learning\u002Fneural-networks\u002Fbackpropagation":16644,"\u002Fdeep-learning\u002Fneural-networks\u002Floss-functions-and-output-units":16648,"\u002Fdeep-learning\u002Foptimization\u002Fgradient-descent-and-sgd":16652,"\u002Fdeep-learning\u002Foptimization\u002Fmomentum-and-adaptive-methods":16657,"\u002Fdeep-learning\u002Foptimization\u002Finitialization":16661,"\u002Fdeep-learning\u002Foptimization\u002Fthe-optimization-landscape":16665,"\u002Fdeep-learning\u002Foptimization\u002Fsecond-order-and-approximate-methods":16669,"\u002Fdeep-learning\u002Fregularization\u002Fregularization-overview":16673,"\u002Fdeep-learning\u002Fregularization\u002Fdropout-and-data-augmentation":16678,"\u002Fdeep-learning\u002Fregularization\u002Fearly-stopping-and-parameter-sharing":16682,"\u002Fdeep-learning\u002Fregularization\u002Fnormalization":16686,"\u002Fdeep-learning\u002Farchitectures\u002Fconvolutional-networks":16690,"\u002Fdeep-learning\u002Farchitectures\u002Fcnn-architectures":16695,"\u002Fdeep-learning\u002Farchitectures\u002Frecurrent-networks":16699,"\u002Fdeep-learning\u002Farchitectures\u002Flstm-and-gru":16702,"\u002Fdeep-learning\u002Farchitectures\u002Fattention-and-transformers":16704,"\u002Fdeep-learning\u002Farchitectures\u002Fthe-transformer-architecture":16708,"\u002Fdeep-learning\u002Farchitectures\u002Ftransformers-in-practice":16712,"\u002Fdeep-learning\u002Farchitectures\u002Fgraph-neural-networks":16716,"\u002Fdeep-learning\u002Farchitectures\u002Fstate-space-models":16720,"\u002Fdeep-learning\u002Ftheory\u002Fgeneralization-theory":16724,"\u002Fdeep-learning\u002Ftheory\u002Fadversarial-robustness":16729,"\u002Fdeep-learning\u002Ftheory\u002Fadversarial-defenses":16733,"\u002Fdeep-learning\u002Ftheory\u002Fbayesian-and-ensemble-methods":16737,"\u002Fdeep-learning\u002Ftheory\u002Fdeep-equilibrium-models":16741,"\u002Fdeep-learning\u002Fgenerative-models\u002Flinear-factor-models":16745,"\u002Fdeep-learning\u002Fgenerative-models\u002Fautoencoders":16750,"\u002Fdeep-learning\u002Fgenerative-models\u002Fvariational-autoencoders":16754,"\u002Fdeep-learning\u002Fgenerative-models\u002Fgenerative-adversarial-networks":16758,"\u002Fdeep-learning\u002Fgenerative-models\u002Fautoregressive-and-normalizing-flows":16762,"\u002Fdeep-learning\u002Fgenerative-models\u002Fenergy-based-and-boltzmann-machines":16766,"\u002Fdeep-learning\u002Fgenerative-models\u002Fdiffusion-and-score-based-models":16770,"\u002Fdeep-learning\u002Fprobabilistic-methods\u002Fstructured-probabilistic-models":16774,"\u002Fdeep-learning\u002Fprobabilistic-methods\u002Fmonte-carlo-and-mcmc":16779,"\u002Fdeep-learning\u002Fprobabilistic-methods\u002Fapproximate-inference":16783,"\u002Fdeep-learning\u002Fpractical\u002Fpractical-methodology":16787,"\u002Fdeep-learning\u002Fpractical\u002Fhyperparameters-and-debugging":16792,"\u002Fdeep-learning\u002Fpractical\u002Frepresentation-learning":16796,"\u002Fdeep-learning\u002Fpractical\u002Ftransfer-learning":16800,"\u002Fdeep-learning\u002Fpractical\u002Fapplications":16804,"\u002Fdeep-learning\u002Fpractical\u002Fmodel-compression-and-distillation":16808,"\u002Fdeep-learning\u002Fpractical\u002Fmeta-learning-and-few-shot":16812,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Flarge-language-models":16816,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fscaling-inference-and-alignment":16821,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fseq2seq-pretraining-and-bart":16825,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Ftext-to-text-transfer-and-conditional-generation":16829,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fspeech-and-audio-models":16833,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fself-supervised-speech-and-synthesis":16837,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fai-agents":16841,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fagent-memory-retrieval-and-orchestration":16845,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fmixture-of-experts":16849,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fmultimodal-models":16853,"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Ffusion-and-vision-language-models":16857,"\u002Fdeep-learning\u002Freinforcement-learning\u002Ffoundations-of-reinforcement-learning":16861,"\u002Fdeep-learning\u002Freinforcement-learning\u002Fmodel-free-prediction-and-control":16866,"\u002Fdeep-learning\u002Freinforcement-learning\u002Fdeep-q-networks":16870,"\u002Fdeep-learning\u002Freinforcement-learning\u002Fpolicy-gradients-and-actor-critic":16874,"\u002Fdeep-learning\u002Freinforcement-learning\u002Frl-from-human-feedback":16878,"\u002Fdeep-learning":16882,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fequilibrium-state-variables-zeroth-law":16885,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Ffirst-law-heat-and-work":16889,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fsecond-law-entropy-and-the-carnot-bound":16893,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fthermodynamic-potentials-and-maxwell-relations":16897,"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fstability-response-functions-and-the-third-law":16901,"\u002Fstatistical-mechanics\u002Ffoundations\u002Fclassical-statistics-and-equipartition":16905,"\u002Fstatistical-mechanics\u002Ffoundations\u002Fphase-space-and-liouvilles-theorem":16910,"\u002Fstatistical-mechanics\u002Ffoundations\u002Fensembles-and-the-equal-probability-postulate":16914,"\u002Fstatistical-mechanics\u002Ffoundations\u002Fstatistical-entropy-boltzmann-and-gibbs":16918,"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Fmicrocanonical-ensemble-and-entropy":16922,"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Fequilibrium-conditions-temperature-pressure-chemical-potential":16927,"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Fideal-gas-phase-space-and-the-sackur-tetrode-entropy":16931,"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Ftwo-state-systems-paramagnets-and-negative-temperature":16935,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fcanonical-ensemble-and-the-boltzmann-distribution":16939,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fpartition-function-and-the-helmholtz-free-energy":16944,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fenergy-fluctuations-and-ensemble-equivalence":16948,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fthe-einstein-solid-and-harmonic-systems":16952,"\u002Fstatistical-mechanics\u002Fcanonical\u002Fparamagnetism-and-the-schottky-anomaly":16956,"\u002Fstatistical-mechanics\u002Fclassical-gas\u002Fideal-gas-partition-function-and-the-gibbs-paradox":16960,"\u002Fstatistical-mechanics\u002Fclassical-gas\u002Fequipartition-and-the-virial-theorem":16965,"\u002Fstatistical-mechanics\u002Fclassical-gas\u002Fmolecular-gases-rotation-and-vibration":16969,"\u002Fstatistical-mechanics\u002Fgrand-canonical\u002Fgrand-canonical-ensemble-and-the-grand-partition-function":16973,"\u002Fstatistical-mechanics\u002Fgrand-canonical\u002Fchemical-potential-fugacity-and-number-fluctuations":16978,"\u002Fstatistical-mechanics\u002Fgrand-canonical\u002Fensemble-summary-and-the-thermodynamic-web":16982,"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fquantum-statistics-bose-einstein-and-fermi-dirac":16986,"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fderiving-the-quantum-distributions":16991,"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fthe-classical-limit-and-quantum-concentration":16995,"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fideal-quantum-gases-general-framework":16999,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fbose-einstein-condensation-and-the-fermion-gas":17003,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fthe-photon-gas-and-plancks-radiation-law":17008,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fblackbody-thermodynamics-and-radiation-pressure":17012,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fphonons-and-the-debye-model":17016,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fbose-einstein-condensation-derived":17020,"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fthermodynamics-of-the-bose-gas-and-superfluidity":17024,"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fthe-ideal-fermi-gas-at-zero-temperature":17028,"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fsommerfeld-expansion-and-electrons-in-metals":17033,"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fwhite-dwarfs-and-the-chandrasekhar-limit":17037,"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fneutron-stars-and-nuclear-matter":17041,"\u002Fstatistical-mechanics\u002Finteractions\u002Fthe-cluster-expansion-and-virial-coefficients":17045,"\u002Fstatistical-mechanics\u002Finteractions\u002Fthe-van-der-waals-gas-and-liquid-gas-coexistence":17050,"\u002Fstatistical-mechanics\u002Finteractions\u002Fquantum-gases-with-interactions-and-exchange":17054,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fphases-coexistence-and-classification":17058,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fthe-ising-model-and-exact-solutions":17063,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fmean-field-theory-and-the-weiss-model":17067,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fcritical-exponents-and-landau-theory":17071,"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fthe-renormalization-group-idea":17075,"\u002Fstatistical-mechanics\u002Ffluctuations\u002Fthermodynamic-fluctuations-and-response":17079,"\u002Fstatistical-mechanics\u002Ffluctuations\u002Fbrownian-motion-and-the-langevin-equation":17084,"\u002Fstatistical-mechanics\u002Ffluctuations\u002Flinear-response-and-the-fluctuation-dissipation-theorem":17088,"\u002Fstatistical-mechanics":17092,"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fbonding-mechanisms":17095,"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fmolecular-orbitals-and-h2-plus":17100,"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fhydrogen-molecule-and-exchange":17104,"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fvan-der-waals-forces":17108,"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Frotational-vibrational-spectra":17112,"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Fanharmonicity-and-rovibrational-structure":17117,"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Framan-and-electronic-bands":17121,"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Flasers-and-masers":17125,"\u002Fcondensed-matter\u002Fcrystal-structure\u002Fstructure-of-solids":17129,"\u002Fcondensed-matter\u002Fcrystal-structure\u002Fbravais-lattices-and-crystal-systems":17134,"\u002Fcondensed-matter\u002Fcrystal-structure\u002Freciprocal-lattice-and-brillouin-zones":17138,"\u002Fcondensed-matter\u002Fcrystal-structure\u002Fdiffraction-and-structure-factors":17142,"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fphonon-dispersion":17146,"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fphonons-quantization-and-dos":17151,"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fdebye-einstein-heat-capacity":17155,"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fanharmonicity-and-thermal-transport":17159,"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Ffree-electron-gas-and-conduction":17163,"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Fsommerfeld-model-and-heat-capacity":17168,"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Ftransport-and-the-hall-effect":17172,"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Fscreening-and-plasmons":17176,"\u002Fcondensed-matter\u002Fband-theory\u002Fblochs-theorem-and-energy-bands":17180,"\u002Fcondensed-matter\u002Fband-theory\u002Fnearly-free-electron-model":17185,"\u002Fcondensed-matter\u002Fband-theory\u002Ftight-binding-method":17189,"\u002Fcondensed-matter\u002Fband-theory\u002Ffermi-surfaces-and-semiclassical-dynamics":17193,"\u002Fcondensed-matter\u002Fsemiconductors\u002Fsemiconductor-bands-and-junctions":17197,"\u002Fcondensed-matter\u002Fsemiconductors\u002Fintrinsic-and-extrinsic-semiconductors":17202,"\u002Fcondensed-matter\u002Fsemiconductors\u002Fcarrier-transport-and-recombination":17206,"\u002Fcondensed-matter\u002Fsemiconductors\u002Fthe-pn-junction":17210,"\u002Fcondensed-matter\u002Fsemiconductors\u002Ftransistors-and-optoelectronics":17214,"\u002Fcondensed-matter\u002Fdielectrics-and-ferroelectrics\u002Fdielectrics-and-polarization":17218,"\u002Fcondensed-matter\u002Fdielectrics-and-ferroelectrics\u002Fferroelectrics-and-piezoelectrics":17223,"\u002Fcondensed-matter\u002Fmagnetism\u002Fdiamagnetism-and-paramagnetism":17227,"\u002Fcondensed-matter\u002Fmagnetism\u002Fexchange-and-ferromagnetism":17232,"\u002Fcondensed-matter\u002Fmagnetism\u002Fantiferromagnetism-and-domains":17236,"\u002Fcondensed-matter\u002Fmagnetism\u002Fspin-waves-and-magnons":17240,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fsuperconductivity-phenomenology":17244,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Flondon-theory-and-the-meissner-effect":17249,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fginzburg-landau-theory":17253,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fbcs-theory":17257,"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fjosephson-and-high-tc":17261,"\u002Fcondensed-matter\u002Fnanostructures\u002Fquantum-wells-wires-and-dots":17265,"\u002Fcondensed-matter\u002Fnanostructures\u002Finteger-quantum-hall-effect":17270,"\u002Fcondensed-matter\u002Fnanostructures\u002Ffractional-quantum-hall-and-topology":17274,"\u002Fcondensed-matter\u002Fnanostructures\u002Fgraphene-and-dirac-materials":17278,"\u002Fcondensed-matter":17282,"\u002Flogic\u002Ffoundations\u002Flogic-as-a-mathematical-model":17285,"\u002Flogic\u002Fsentential-logic\u002Fformal-languages-and-well-formed-formulas":17289,"\u002Flogic\u002Fsentential-logic\u002Ftruth-assignments-and-tautologies":17294,"\u002Flogic\u002Fsentential-logic\u002Funique-readability-and-parsing":17298,"\u002Flogic\u002Fsentential-logic\u002Finduction-and-recursion":17302,"\u002Flogic\u002Fsentential-logic\u002Fexpressive-completeness-and-normal-forms":17306,"\u002Flogic\u002Fsentential-logic\u002Fboolean-circuits":17310,"\u002Flogic\u002Fsentential-logic\u002Fcompactness-and-effectiveness":17314,"\u002Flogic\u002Ffirst-order-languages\u002Ffirst-order-languages":17318,"\u002Flogic\u002Ffirst-order-languages\u002Fstructures-truth-and-satisfaction":17323,"\u002Flogic\u002Ffirst-order-languages\u002Fdefinability-and-elementary-equivalence":17327,"\u002Flogic\u002Ffirst-order-languages\u002Fterms-substitution-and-parsing":17331,"\u002Flogic\u002Fdeductive-calculus\u002Fa-deductive-calculus":17335,"\u002Flogic\u002Fdeductive-calculus\u002Fdeduction-theorem-and-derived-rules":17340,"\u002Flogic\u002Fdeductive-calculus\u002Fsoundness":17344,"\u002Flogic\u002Fdeductive-calculus\u002Fcompleteness-and-consistency":17348,"\u002Flogic\u002Fmodels-and-theories\u002Fcompactness-and-lowenheim-skolem":17352,"\u002Flogic\u002Fmodels-and-theories\u002Ftheories-elementary-classes-and-categoricity":17357,"\u002Flogic\u002Fmodels-and-theories\u002Finterpretations-between-theories":17361,"\u002Flogic\u002Fmodels-and-theories\u002Fnonstandard-analysis":17365,"\u002Flogic\u002Farithmetic-and-definability\u002Fdefinability-in-arithmetic":17369,"\u002Flogic\u002Farithmetic-and-definability\u002Fnatural-numbers-with-successor":17374,"\u002Flogic\u002Farithmetic-and-definability\u002Fpresburger-and-reducts":17378,"\u002Flogic\u002Farithmetic-and-definability\u002Fa-subtheory-and-representability":17382,"\u002Flogic\u002Fincompleteness\u002Farithmetization-of-syntax":17386,"\u002Flogic\u002Fincompleteness\u002Fincompleteness-and-undecidability":17391,"\u002Flogic\u002Fincompleteness\u002Fsecond-incompleteness-theorem":17395,"\u002Flogic\u002Fcomputability-and-representability\u002Frecursive-functions":17399,"\u002Flogic\u002Fcomputability-and-representability\u002Frepresenting-exponentiation":17404,"\u002Flogic\u002Fsecond-order-logic\u002Fsecond-order-languages":17408,"\u002Flogic\u002Fsecond-order-logic\u002Fskolem-functions-and-many-sorted-logic":17413,"\u002Flogic\u002Fsecond-order-logic\u002Fgeneral-structures":17417,"\u002Flogic":17421,"\u002Freinforcement-learning\u002Ffoundations\u002Fwhat-is-reinforcement-learning":17424,"\u002Freinforcement-learning\u002Ffoundations\u002Fa-brief-history-of-rl":17428,"\u002Freinforcement-learning\u002Ffoundations\u002Fmulti-armed-bandits":17432,"\u002Freinforcement-learning\u002Ffoundations\u002Fbandit-exploration-algorithms":17436,"\u002Freinforcement-learning\u002Ffoundations\u002Fmarkov-decision-processes":17440,"\u002Freinforcement-learning\u002Ffoundations\u002Fvalue-functions-and-optimality":17444,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fdynamic-programming":17448,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fdp-async-and-gpi":17452,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fmonte-carlo-methods":17456,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fmonte-carlo-off-policy":17460,"\u002Freinforcement-learning\u002Ftabular-methods\u002Ftemporal-difference-learning":17464,"\u002Freinforcement-learning\u002Ftabular-methods\u002Ftd-control-sarsa-and-q-learning":17468,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fn-step-bootstrapping":17472,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fn-step-off-policy-methods":17476,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fplanning-and-learning":17480,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fplanning-focusing-and-decision-time":17484,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fdecision-time-planning":17488,"\u002Freinforcement-learning\u002Ftabular-methods\u002Fmonte-carlo-tree-search":17492,"\u002Freinforcement-learning\u002Fapproximation\u002Fon-policy-prediction":17496,"\u002Freinforcement-learning\u002Fapproximation\u002Ffeature-construction-and-nonlinear":17501,"\u002Freinforcement-learning\u002Fapproximation\u002Fon-policy-control":17505,"\u002Freinforcement-learning\u002Fapproximation\u002Faverage-reward-control":17509,"\u002Freinforcement-learning\u002Fapproximation\u002Foff-policy-and-the-deadly-triad":17513,"\u002Freinforcement-learning\u002Fapproximation\u002Fbellman-error-and-gradient-td":17517,"\u002Freinforcement-learning\u002Fapproximation\u002Feligibility-traces":17521,"\u002Freinforcement-learning\u002Fapproximation\u002Ftrue-online-and-sarsa-lambda":17525,"\u002Freinforcement-learning\u002Fapproximation\u002Fpolicy-gradient-methods":17529,"\u002Freinforcement-learning\u002Fapproximation\u002Factor-critic-and-continuous-actions":17533,"\u002Freinforcement-learning\u002Fapproximation\u002Fleast-squares-and-memory-based-methods":17537,"\u002Freinforcement-learning\u002Fapproximation\u002Fmemory-and-kernel-methods":17541,"\u002Freinforcement-learning\u002Fapproximation\u002Foff-policy-eligibility-traces":17545,"\u002Freinforcement-learning\u002Fapproximation\u002Fstable-off-policy-traces":17549,"\u002Freinforcement-learning\u002Fdeep-rl\u002Fdeep-q-networks":17553,"\u002Freinforcement-learning\u002Fdeep-rl\u002Fdqn-improvements":17557,"\u002Freinforcement-learning\u002Fdeep-rl\u002Factor-critic-and-ppo":17561,"\u002Freinforcement-learning\u002Fdeep-rl\u002Fppo-and-continuous-control":17565,"\u002Freinforcement-learning\u002Fdeep-rl\u002Fcase-studies":17569,"\u002Freinforcement-learning\u002Fdeep-rl\u002Frl-beyond-games":17573,"\u002Freinforcement-learning\u002Fdeep-rl\u002Ffrontiers":17577,"\u002Freinforcement-learning\u002Fdeep-rl\u002Freward-design-and-open-problems":17581,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fdistributional-and-rainbow":17585,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fdistributional-and-rainbow-part-2":17590,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fcontinuous-control":17594,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fcontinuous-control-part-2":17598,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmodel-based-rl":17602,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmodel-based-rl-part-2":17606,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fexploration":17610,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fexploration-part-2":17614,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Foffline-rl":17618,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Foffline-rl-part-2":17622,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fimitation-and-inverse-rl":17626,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fimitation-and-inverse-rl-part-2":17630,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmulti-agent-rl":17634,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmulti-agent-rl-part-2":17638,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fhierarchical-rl":17642,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fhierarchical-rl-part-2":17646,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Frlhf-and-language-models":17650,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fpartial-observability-pomdps":17654,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fpartial-observability-pomdps-part-2":17658,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fsafe-and-constrained-rl":17662,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fsafe-and-constrained-rl-part-2":17666,"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmeta-rl-and-generalization":17670,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fpsychology-of-reinforcement":17674,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Finstrumental-conditioning-and-control":17679,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fdopamine-and-td-error":17683,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fdopamine-in-the-brain":17687,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fanimal-learning-and-cognition":17691,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fcognitive-maps-and-planning":17695,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fneuroscience-of-reinforcement":17699,"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fseveral-learning-systems":17703,"\u002Freinforcement-learning":17707,"\u002Fartificial-intelligence\u002Ffoundations\u002Fwhat-is-ai":17709,"\u002Fartificial-intelligence\u002Ffoundations\u002Ffoundations-of-ai":17713,"\u002Fartificial-intelligence\u002Ffoundations\u002Fintelligent-agents":17717,"\u002Fartificial-intelligence\u002Ffoundations\u002Fagent-architectures":17721,"\u002Fartificial-intelligence\u002Fsearch\u002Funinformed-search":17725,"\u002Fartificial-intelligence\u002Fsearch\u002Fsearch-strategies-compared":17730,"\u002Fartificial-intelligence\u002Fsearch\u002Finformed-search":17734,"\u002Fartificial-intelligence\u002Fsearch\u002Fheuristic-functions":17738,"\u002Fartificial-intelligence\u002Fsearch\u002Flocal-search":17742,"\u002Fartificial-intelligence\u002Fsearch\u002Fpopulation-and-continuous-search":17746,"\u002Fartificial-intelligence\u002Fsearch\u002Fadversarial-search":17750,"\u002Fartificial-intelligence\u002Fsearch\u002Fgames-of-chance-and-imperfect-information":17754,"\u002Fartificial-intelligence\u002Fsearch\u002Fconstraint-satisfaction":17758,"\u002Fartificial-intelligence\u002Fsearch\u002Fcsp-search-and-structure":17762,"\u002Fartificial-intelligence\u002Fsearch\u002Fsearch-under-uncertainty":17766,"\u002Fartificial-intelligence\u002Fsearch\u002Fbelief-state-and-online-search":17770,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fpropositional-logic":17774,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fpropositional-inference":17779,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Ffirst-order-logic":17783,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Ffirst-order-logic-in-use":17787,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Finference-and-resolution":17791,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Ffirst-order-resolution":17795,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fclassical-planning":17799,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fplanning-graphs-and-graphplan":17803,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fplanning-in-the-real-world":17807,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fplanning-under-uncertainty":17811,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fknowledge-representation":17815,"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Freasoning-systems-and-defaults":17819,"\u002Fartificial-intelligence\u002Funcertainty\u002Fprobability-and-bayes":17823,"\u002Fartificial-intelligence\u002Funcertainty\u002Fbayes-rule-and-naive-bayes":17828,"\u002Fartificial-intelligence\u002Funcertainty\u002Fbayesian-networks":17832,"\u002Fartificial-intelligence\u002Funcertainty\u002Finference-in-bayesian-networks":17836,"\u002Fartificial-intelligence\u002Funcertainty\u002Freasoning-over-time":17840,"\u002Fartificial-intelligence\u002Funcertainty\u002Ftracking-and-data-association":17844,"\u002Fartificial-intelligence\u002Funcertainty\u002Fmaking-decisions":17848,"\u002Fartificial-intelligence\u002Funcertainty\u002Fmarkov-decision-processes":17852,"\u002Fartificial-intelligence\u002Funcertainty\u002Fdecision-networks-and-game-theory":17855,"\u002Fartificial-intelligence\u002Funcertainty\u002Fgame-theory-and-mechanism-design":17859,"\u002Fartificial-intelligence\u002Flearning\u002Flearning-from-examples":17863,"\u002Fartificial-intelligence\u002Flearning\u002Ftheory-and-model-families":17868,"\u002Fartificial-intelligence\u002Flearning\u002Fprobabilistic-learning":17872,"\u002Fartificial-intelligence\u002Flearning\u002Fexpectation-maximization":17876,"\u002Fartificial-intelligence\u002Flearning\u002Freinforcement-learning":17880,"\u002Fartificial-intelligence\u002Flearning\u002Fgeneralization-and-policy-search":17883,"\u002Fartificial-intelligence\u002Flearning\u002Fknowledge-in-learning":17887,"\u002Fartificial-intelligence\u002Flearning\u002Fknowledge-based-learning-methods":17891,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fvision-and-perception":17895,"\u002Fartificial-intelligence\u002Ffrontiers\u002Freconstructing-the-3d-world":17900,"\u002Fartificial-intelligence\u002Ffrontiers\u002Frobotics":17904,"\u002Fartificial-intelligence\u002Ffrontiers\u002Frobot-planning-and-control":17908,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fnatural-language-in-ai":17912,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fnlp-grammar-translation-and-speech":17916,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fphilosophy-and-future":17920,"\u002Fartificial-intelligence\u002Ffrontiers\u002Fai-ethics-and-future":17924,"\u002Fartificial-intelligence":17928,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-constituents-nuclide-chart":17931,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-size-charge-distributions":17936,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-masses-binding-energy":17940,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fsemi-empirical-mass-formula":17944,"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-moments-multipoles":17948,"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fnuclear-force-shell-overview":17952,"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fthe-deuteron":17957,"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fnucleon-nucleon-scattering":17961,"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fmeson-theory-isospin":17965,"\u002Fnuclear-physics\u002Fnuclear-models\u002Ffermi-gas-model":17969,"\u002Fnuclear-physics\u002Fnuclear-models\u002Fliquid-drop-collective-coordinates":17974,"\u002Fnuclear-physics\u002Fnuclear-models\u002Fshell-model-single-particle":17978,"\u002Fnuclear-physics\u002Fnuclear-models\u002Fcollective-model-rotations-vibrations":17982,"\u002Fnuclear-physics\u002Fradioactive-decay\u002Fdecay-law-modes":17986,"\u002Fnuclear-physics\u002Fradioactive-decay\u002Fdecay-kinetics-equilibrium":17991,"\u002Fnuclear-physics\u002Falpha-decay\u002Falpha-decay-gamow-theory":17995,"\u002Fnuclear-physics\u002Falpha-decay\u002Falpha-fine-structure-hindrance":18000,"\u002Fnuclear-physics\u002Fbeta-decay\u002Fbeta-decay-energetics-neutrino":18004,"\u002Fnuclear-physics\u002Fbeta-decay\u002Ffermi-theory-beta-decay":18009,"\u002Fnuclear-physics\u002Fbeta-decay\u002Fweak-interaction-parity-violation":18013,"\u002Fnuclear-physics\u002Fbeta-decay\u002Fdouble-beta-decay-neutrino-mass":18017,"\u002Fnuclear-physics\u002Fgamma-decay\u002Fgamma-multipole-radiation":18021,"\u002Fnuclear-physics\u002Fgamma-decay\u002Finternal-conversion-isomers":18026,"\u002Fnuclear-physics\u002Fgamma-decay\u002Fangular-correlations-mossbauer":18030,"\u002Fnuclear-physics\u002Fnuclear-reactions\u002Freaction-kinematics-cross-sections":18034,"\u002Fnuclear-physics\u002Fnuclear-reactions\u002Fcompound-nucleus-resonances":18039,"\u002Fnuclear-physics\u002Fnuclear-reactions\u002Fdirect-reactions-optical-model":18043,"\u002Fnuclear-physics\u002Ffission\u002Ffission-barrier-dynamics":18047,"\u002Fnuclear-physics\u002Ffission\u002Fchain-reactions-reactor-physics":18052,"\u002Fnuclear-physics\u002Ffusion-nucleosynthesis\u002Ffusion-reactions-confinement":18056,"\u002Fnuclear-physics\u002Ffusion-nucleosynthesis\u002Fstellar-nucleosynthesis":18061,"\u002Fnuclear-physics\u002Ffusion-nucleosynthesis\u002Fbig-bang-nucleosynthesis":18065,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fcharged-particle-stopping-power":18069,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fphoton-neutron-interactions":18074,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fradiation-detectors":18078,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fdosimetry-radiation-biology":18082,"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fnuclear-applications-dating-medicine":18086,"\u002Fnuclear-physics":18090,"\u002Fnatural-language-processing\u002Ffoundations\u002Fwhat-is-nlp":18093,"\u002Fnatural-language-processing\u002Ffoundations\u002Fregex-and-text-normalization":18097,"\u002Fnatural-language-processing\u002Ffoundations\u002Fminimum-edit-distance":18101,"\u002Fnatural-language-processing\u002Ffoundations\u002Fn-gram-language-models":18105,"\u002Fnatural-language-processing\u002Ffoundations\u002Fsmoothing-and-backoff":18109,"\u002Fnatural-language-processing\u002Fclassification\u002Fnaive-bayes-and-sentiment":18113,"\u002Fnatural-language-processing\u002Fclassification\u002Fevaluating-classifiers":18118,"\u002Fnatural-language-processing\u002Fclassification\u002Flogistic-regression":18122,"\u002Fnatural-language-processing\u002Fclassification\u002Fsentiment-and-affect-lexicons":18126,"\u002Fnatural-language-processing\u002Fsemantics\u002Fvector-semantics-and-embeddings":18130,"\u002Fnatural-language-processing\u002Fsemantics\u002Fstatic-word-embeddings":18135,"\u002Fnatural-language-processing\u002Fsemantics\u002Fneural-language-models":18139,"\u002Fnatural-language-processing\u002Fsequences\u002Fsequence-labeling":18142,"\u002Fnatural-language-processing\u002Fsequences\u002Fcrfs-and-neural-taggers":18145,"\u002Fnatural-language-processing\u002Fsequences\u002Frnns-and-lstms":18149,"\u002Fnatural-language-processing\u002Ftransformers\u002Ftransformers-and-attention":18150,"\u002Fnatural-language-processing\u002Ftransformers\u002Fthe-transformer-architecture":18153,"\u002Fnatural-language-processing\u002Ftransformers\u002Flarge-language-models":18156,"\u002Fnatural-language-processing\u002Ftransformers\u002Fllm-pretraining-and-scaling":18159,"\u002Fnatural-language-processing\u002Ftransformers\u002Ffine-tuning-and-prompting":18163,"\u002Fnatural-language-processing\u002Ftransformers\u002Fprompting-and-alignment":18167,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fconstituency-parsing":18171,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcky-scoring-and-evaluation":18176,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fdependency-parsing":18180,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fgraph-based-and-neural-dependency-parsing":18184,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fword-senses-and-wsd":18188,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fwsd-in-practice-and-induction":18192,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fsemantic-roles-and-information-extraction":18196,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Frelations-events-and-templates":18200,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcoreference-and-discourse":18204,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcoherence-and-discourse-structure":18208,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Flogical-semantics":18212,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcompositional-semantics-and-description-logics":18216,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fsemantic-parsing":18220,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fneural-semantic-parsing":18224,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Finformation-extraction":18228,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Ftimes-events-and-templates":18232,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fdiscourse-coherence":18236,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fentity-based-and-global-coherence":18240,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fconstituency-grammars":18244,"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Ftreebanks-and-lexicalized-grammars":18248,"\u002Fnatural-language-processing\u002Fapplications\u002Fmachine-translation":18252,"\u002Fnatural-language-processing\u002Fapplications\u002Fmachine-translation-decoding-and-evaluation":18256,"\u002Fnatural-language-processing\u002Fapplications\u002Fquestion-answering":18260,"\u002Fnatural-language-processing\u002Fapplications\u002Fquestion-answering-knowledge-and-llms":18264,"\u002Fnatural-language-processing\u002Fapplications\u002Fdialogue-and-chatbots":18268,"\u002Fnatural-language-processing\u002Fapplications\u002Fdialogue-systems-and-assistants":18272,"\u002Fnatural-language-processing\u002Fapplications\u002Ftext-summarization":18276,"\u002Fnatural-language-processing\u002Fapplications\u002Fabstractive-summarization-and-evaluation":18280,"\u002Fnatural-language-processing\u002Fspeech\u002Fphonetics":18284,"\u002Fnatural-language-processing\u002Fspeech\u002Facoustic-phonetics":18289,"\u002Fnatural-language-processing\u002Fspeech\u002Fautomatic-speech-recognition":18293,"\u002Fnatural-language-processing\u002Fspeech\u002Fasr-evaluation-and-applications":18297,"\u002Fnatural-language-processing":18301,"\u002Fparticle-physics\u002Ffoundations\u002Fhistorical-overview-particle-zoo":18304,"\u002Fparticle-physics\u002Ffoundations\u002Fparticle-physics-basic-concepts":18308,"\u002Fparticle-physics\u002Ffoundations\u002Ffundamental-interactions-force-carriers":18312,"\u002Fparticle-physics\u002Funits-kinematics\u002Fnatural-units-and-scales":18316,"\u002Fparticle-physics\u002Funits-kinematics\u002Ffour-vectors-invariant-mass":18321,"\u002Fparticle-physics\u002Funits-kinematics\u002Fdecay-scattering-kinematics-mandelstam":18325,"\u002Fparticle-physics\u002Funits-kinematics\u002Fcross-sections-golden-rule":18329,"\u002Fparticle-physics\u002Fsymmetries\u002Fconservation-laws-symmetries":18333,"\u002Fparticle-physics\u002Fsymmetries\u002Fdiscrete-symmetries-cpt":18338,"\u002Fparticle-physics\u002Fsymmetries\u002Fparity-violation-weak":18342,"\u002Fparticle-physics\u002Fsymmetries\u002Fsu2-su3-flavor-symmetry":18346,"\u002Fparticle-physics\u002Fquark-model\u002Feightfold-way-su3":18350,"\u002Fparticle-physics\u002Fquark-model\u002Fmeson-spectroscopy":18355,"\u002Fparticle-physics\u002Fquark-model\u002Fbaryon-spectroscopy":18359,"\u002Fparticle-physics\u002Fquark-model\u002Fcolor-confinement-exotics":18363,"\u002Fparticle-physics\u002Frelativistic-wave-equations\u002Fklein-gordon-equation":18367,"\u002Fparticle-physics\u002Frelativistic-wave-equations\u002Fdirac-equation-spinors":18372,"\u002Fparticle-physics\u002Frelativistic-wave-equations\u002Fantiparticles-hole-theory":18376,"\u002Fparticle-physics\u002Fqed\u002Ffeynman-rules-qed":18380,"\u002Fparticle-physics\u002Fqed\u002Fqed-tree-processes":18385,"\u002Fparticle-physics\u002Fqed\u002Frenormalization-running-coupling":18389,"\u002Fparticle-physics\u002Fqed\u002Felectron-g-2":18393,"\u002Fparticle-physics\u002Fweak-interaction\u002Fva-structure-weak":18397,"\u002Fparticle-physics\u002Fweak-interaction\u002Fw-z-bosons-decays":18402,"\u002Fparticle-physics\u002Fweak-interaction\u002Fckm-matrix":18406,"\u002Fparticle-physics\u002Fweak-interaction\u002Fcp-violation-kaons-b-mesons":18410,"\u002Fparticle-physics\u002Fqcd\u002Fcolor-su3-gluons":18414,"\u002Fparticle-physics\u002Fqcd\u002Fasymptotic-freedom-confinement":18419,"\u002Fparticle-physics\u002Fqcd\u002Fdeep-inelastic-scattering-partons":18423,"\u002Fparticle-physics\u002Fqcd\u002Fjets-hadronization":18427,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Felectroweak-su2-u1":18431,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fspontaneous-symmetry-breaking":18436,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fhiggs-mechanism":18440,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fhiggs-boson-discovery":18444,"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fstandard-model":18448,"\u002Fparticle-physics\u002Fneutrinos\u002Fneutrino-oscillations":18452,"\u002Fparticle-physics\u002Fneutrinos\u002Fneutrino-mass-pmns":18457,"\u002Fparticle-physics\u002Fneutrinos\u002Fdirac-majorana-experiments":18461,"\u002Fparticle-physics\u002Fexperiment\u002Faccelerators-luminosity":18465,"\u002Fparticle-physics\u002Fexperiment\u002Fdetectors-subsystems":18470,"\u002Fparticle-physics\u002Fexperiment\u002Fhow-discoveries-are-made":18474,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fbeyond-standard-model":18478,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fgrand-unified-theories":18482,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fsupersymmetry":18486,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fhierarchy-problem-naturalness":18490,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fdark-matter-candidates":18494,"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fmatter-antimatter-open-questions":18498,"\u002Fparticle-physics":18502,"\u002Fastrophysics-cosmology\u002Forientation\u002Fthe-sun-and-stars":18505,"\u002Fastrophysics-cosmology\u002Forientation\u002Fstellar-death-final-states":18510,"\u002Fastrophysics-cosmology\u002Forientation\u002Fgalaxies-and-cosmology":18514,"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Fmagnitudes-fluxes-and-the-distance-modulus":18518,"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Fstellar-spectra-and-spectral-classification":18523,"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Ftelescopes-and-detectors-across-the-spectrum":18527,"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Fthe-cosmic-distance-ladder":18531,"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fblackbody-radiation-and-specific-intensity":18535,"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fradiative-transfer-and-the-transfer-equation":18540,"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fspectral-line-formation-and-broadening":18544,"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fopacity-and-the-rosseland-mean":18548,"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fhydrostatic-equilibrium-and-the-virial-theorem":18552,"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fthe-equations-of-stellar-structure":18557,"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fthe-equation-of-state-and-polytropes":18561,"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fthe-standard-solar-model":18565,"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fthermonuclear-reaction-rates-and-the-gamow-peak":18569,"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fhydrogen-burning-pp-chains-and-cno":18574,"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fhelium-burning-and-the-triple-alpha-process":18578,"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fadvanced-burning-and-neutron-capture-nucleosynthesis":18582,"\u002Fastrophysics-cosmology\u002Fism-and-star-formation\u002Fphases-of-the-interstellar-medium":18586,"\u002Fastrophysics-cosmology\u002Fism-and-star-formation\u002Fmolecular-clouds-and-gravitational-collapse":18591,"\u002Fastrophysics-cosmology\u002Fism-and-star-formation\u002Fprotostars-and-the-pre-main-sequence":18595,"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fthe-main-sequence-and-its-structure":18599,"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fpost-main-sequence-low-mass-evolution":18604,"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fthe-evolution-of-massive-stars":18608,"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fstellar-pulsation-and-the-instability-strip":18612,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fwhite-dwarfs-and-the-chandrasekhar-limit":18616,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fcore-collapse-supernovae":18620,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fthermonuclear-supernovae-type-ia":18624,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fneutron-stars-and-pulsars":18628,"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fblack-holes-schwarzschild-and-kerr":18632,"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Fbinary-systems-and-mass-transfer":18636,"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Faccreting-compact-objects":18641,"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Fgravitational-waves-from-inspiraling-binaries":18645,"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Fmultimessenger-astronomy-and-gamma-ray-bursts":18649,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fthe-milky-way":18653,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fgalaxy-morphology-and-classification":18658,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fgalaxy-rotation-curves-and-dark-matter":18662,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Factive-galactic-nuclei-and-supermassive-black-holes":18666,"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fgalaxy-clusters-and-large-scale-structure":18670,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fthe-expanding-universe-and-hubbles-law":18674,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fthe-frw-metric-and-cosmological-redshift":18679,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fthe-friedmann-equations-and-cosmic-dynamics":18683,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fcosmological-models-and-distances":18686,"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fdark-energy-and-the-accelerating-universe":18690,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fthe-thermal-history-of-the-universe":18694,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fbig-bang-nucleosynthesis":18699,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Frecombination-and-the-cosmic-microwave-background":18703,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fcmb-anisotropies-and-cosmological-parameters":18707,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fcosmic-inflation":18711,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fstructure-formation-and-the-growth-of-perturbations":18715,"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fdark-matter-dark-energy-and-open-questions":18719,"\u002Fastrophysics-cosmology":18723,"\u002Fcolophon":18726,"\u002F":18729},{"path":14052,"title":14053,"module":14054,"summary":14055},"\u002Falgorithms\u002Ffoundations\u002Fwhat-is-an-algorithm","What Is an Algorithm?","Foundations","An algorithm is a finite, mechanical recipe that transforms inputs into outputs. We define what counts as an algorithm, how we write one down, and the three things we always ask of it: is it correct, is it fast, and can we prove it.\n",{"path":14057,"title":14058,"module":14054,"summary":14059},"\u002Falgorithms\u002Ffoundations\u002Fproof-techniques","Proof Techniques","An algorithm without a proof is a conjecture. This lesson collects the handful\nof arguments that certify the algorithms in this course — direct proof,\ncontrapositive, contradiction, ordinary and strong induction, construction, and\ndisproof by counterexample — each with a small worked\nexample and a picture. Loop invariants are a form of induction,\nrecursive correctness falls to strong induction, and the classic broken proofs\n(all horses are the same color) show where inductions go wrong.\n",{"path":14061,"title":14062,"module":14054,"summary":14063},"\u002Falgorithms\u002Ffoundations\u002Fasymptotic-analysis","Asymptotic Analysis","We measure an algorithm's running time as a function of its input size, then strip away machine-specific constants and lower-order terms to compare algorithms cleanly. This lesson defines the RAM model and the $O$, $\\Omega$, $\\Theta$, $o$, and $\\omega$ notations, proves the polynomial theorem, and shows how to rank growth rates with the limit test, L'Hôpital, base substitution, and the logarithm identities the arguments lean on.\n",{"path":14065,"title":14066,"module":14054,"summary":14067},"\u002Falgorithms\u002Ffoundations\u002Fgrowth-rates-and-loop-analysis","Growth Rates and Loop Analysis","With the asymptotic notations in hand, we rank the functions that actually arise in running times — from constant to factorial — proving the orderings between rungs, then read the running time of a loop nest straight off the page. Sequential blocks add, nested loops multiply, index scaling gives logarithms; a worked trace and a tour of cache-aware and galactic algorithms close the lesson.\n",{"path":14069,"title":14070,"module":14054,"summary":14071},"\u002Falgorithms\u002Ffoundations\u002Frecurrences","Recurrences and the Master Theorem","Recursive and divide-and-conquer algorithms describe their own running time with a recurrence: $T(n)$ in terms of $T$ on smaller inputs. We solve recurrences three ways — drawing the recursion tree, guessing-and-verifying by induction, and applying the Master Theorem — using merge sort as the running example, then handle unequal splits with Akra–Bazzi.\n",{"path":14073,"title":14074,"module":14054,"summary":14075},"\u002Falgorithms\u002Ffoundations\u002Famortized-analysis","Amortized Analysis","Some operations are occasionally expensive but cheap on average across any\nsequence. Amortized analysis bounds the average cost per operation over a\nworst-case sequence — not an expectation — so a rare costly step is paid for by\nthe many cheap ones around it. This lesson develops the aggregate, accounting,\nand potential methods on dynamic-array doubling, the binary counter, and a\nstack with multipop.\n",{"path":14077,"title":14078,"module":14079,"summary":14080},"\u002Falgorithms\u002Fdivide-and-conquer\u002Fmergesort","Divide and Conquer & Mergesort","Divide & Conquer","Divide and conquer breaks a problem into smaller copies of itself, solves\nthem recursively, and stitches the answers together. We meet the paradigm\nthrough mergesort — its merge step, its loop-invariant proof, and the\nrecursion tree that pins its cost at $\\Theta(n\\log n)$ — then count inversions\nwith the same machinery and distill the whole pattern into the master theorem.\n",{"path":14082,"title":14083,"module":14079,"summary":14084},"\u002Falgorithms\u002Fdivide-and-conquer\u002Fquicksort","Quicksort","Quicksort sorts in place by partitioning around a pivot and recursing on\neach side. We give Lomuto and Hoare partitioning with a correctness\ninvariant, see why a bad pivot costs $\\Theta(n^2)$ while a balanced one gives\n$\\Theta(n\\log n)$, and prove that randomizing the pivot makes the expected\ncost $\\Theta(n\\log n)$ on every input.\n",{"path":14086,"title":14087,"module":14079,"summary":14088},"\u002Falgorithms\u002Fdivide-and-conquer\u002Fselection","Linear-Time Selection","Finding the $k$-th smallest element looks like it should require sorting, but\nit does not. Quickselect adapts quicksort's partition to recurse on just one\nside, achieving expected $O(n)$. The median-of-medians algorithm guarantees a\ngood pivot with the groups-of-five trick, pushing the worst case down to a\nprovable $O(n)$.\n",{"path":14090,"title":14091,"module":14079,"summary":14092},"\u002Falgorithms\u002Fdivide-and-conquer\u002Ffast-multiplication","Fast Multiplication","Grade-school multiplication is $\\Theta(n^2)$, yet divide and conquer beats it.\nKaratsuba multiplies $n$-bit integers with three half-size products instead of\nfour, giving $\\Theta(n^{\\log_2 3})$, and Strassen multiplies matrices with\nseven block products instead of eight, giving $\\Theta(n^{\\log_2 7})$. Both\nspend cheap additions to save an expensive multiplication, and the master\ntheorem quantifies the savings.\n",{"path":14094,"title":14095,"module":14096,"summary":14097},"\u002Falgorithms\u002Fsorting\u002Fheaps-and-heapsort","Heaps and Heapsort","Sorting & Order Statistics","A binary heap is a tree we store flat in an array, with index arithmetic\nstanding in for pointers. We build the max-heap property bottom-up in $O(n)$\ntime, sort in place in $\\Theta(n\\log n)$ by repeatedly extracting the maximum,\nand reuse the same structure to implement a priority queue.\n",{"path":14099,"title":14100,"module":14096,"summary":14101},"\u002Falgorithms\u002Fsorting\u002Fsorting-lower-bounds","Lower Bounds for Comparison Sorting","Every sort we have seen runs in $\\Omega(n\\log n)$, and that is no accident.\nModeling a sort as a decision tree of comparisons, we show any such tree must\nhave $n!$ leaves, forcing height $\\ge \\log_2(n!) = \\Omega(n\\log n)$ — a bound\nno comparison sort beats in the worst case, on average, or with randomness.\n",{"path":14103,"title":14104,"module":14096,"summary":14105},"\u002Falgorithms\u002Fsorting\u002Flinear-time-sorting","Sorting in Linear Time","The $\\Omega(n\\log n)$ barrier only binds algorithms that compare. By instead\nusing keys as array indices we slip past it: counting sort runs in\n$\\Theta(n+k)$ and is stable, radix sort layers it digit by digit, and bucket\nsort averages $\\Theta(n)$ on uniform data. We see exactly when each applies.\n",{"path":14107,"title":14108,"module":14096,"summary":14109},"\u002Falgorithms\u002Fsorting\u002Fexternal-sorting","External Sorting","When the data dwarfs main memory, the cost that matters is no longer\ncomparisons but block transfers to and from disk. External merge sort sorts\nmemory-sized runs, then folds them together with a heap-driven $k$-way merge in\n$\\Theta(\\log_k(N\u002FM))$ passes. Larger fan-out cuts passes; replacement selection\nbuilds longer runs to cut them further.\n",{"path":14111,"title":14112,"module":14113,"summary":14114},"\u002Falgorithms\u002Fdata-structures\u002Felementary-structures","Elementary Data Structures","Data Structures","Every container is built one of two ways: **contiguous** in an array, or\n**linked** through pointers. We trade cache-friendly random access against\n$O(1)$ splicing, derive the **amortized $O(1)$** append of a doubling dynamic\narray, and assemble the two ordered access disciplines — the LIFO **stack** and\nthe FIFO **queue** (with its generalization, the **deque**) — on top of both.\n",{"path":14116,"title":14117,"module":14113,"summary":14118},"\u002Falgorithms\u002Fdata-structures\u002Fhash-tables","Hash Tables","A hash table implements the dictionary — insert, search, delete — in expected\n$O(1)$ time by scattering keys across an array with a hash function. We build\nup from direct addressing, handle collisions by chaining and by open\naddressing, analyze the load factor $\\alpha$, and see how universal hashing\nachieves its expected-time guarantee against every input.\n",{"path":14120,"title":14121,"module":14113,"summary":14122},"\u002Falgorithms\u002Fdata-structures\u002Fbinary-search-trees","Binary Search Trees","A binary search tree keeps keys ordered so that every operation follows a\nsingle root-to-leaf path. We state the BST property, trace search, insert,\nsuccessor, and all three delete cases on concrete trees, prove the inorder\nwalk sorts, and note the drawback — every operation costs $O(h)$, and a\ncarelessly built tree degrades to height $h = \\Theta(n)$, motivating balance.\n",{"path":14124,"title":14125,"module":14113,"summary":14126},"\u002Falgorithms\u002Fdata-structures\u002Favl-trees","AVL Trees","An AVL tree is the first balanced BST: at every node the two subtrees' heights\ndiffer by at most $1$. A Fibonacci-style minimal-node argument forces height\n$h \\le 1.44\\log_2 n = O(\\log n)$, so search, insert, and delete are all\n$O(\\log n)$. Insertion rebalances with at most one of four rotation cases\n(LL, RR, LR, RL); deletion may rotate all the way to the root.\n",{"path":14128,"title":14129,"module":14113,"summary":14130},"\u002Falgorithms\u002Fdata-structures\u002Fbalanced-trees","Balanced Search Trees","An ordinary BST can degrade to height $\\Theta(n)$; balanced search trees\nguarantee $h = O(\\log n)$ by maintaining invariants and repairing them after\nevery update. We meet rotations, the local restructuring primitive, then\nred-black trees, whose color invariants force logarithmic height, and finally\nB-trees, which trade tall-and-thin for short-and-wide to win on disk.\n",{"path":14132,"title":14133,"module":14113,"summary":14134},"\u002Falgorithms\u002Fdata-structures\u002Funion-find","Disjoint Sets (Union-Find)","The disjoint-set data structure tracks a partition of elements into groups,\nanswering \"are these two in the same group?\" and merging groups on demand. A\nforest of parent pointers, sped up by union by rank and path compression,\ndrives every operation to near-constant $O(\\alpha(n))$ amortized time — the\nstructure behind connectivity queries and Kruskal's minimum spanning tree.\n",{"path":14136,"title":14137,"module":14113,"summary":14138},"\u002Falgorithms\u002Fdata-structures\u002Ffenwick-and-segment-trees","Fenwick & Segment Trees","A prefix-sum array answers a range sum in $O(1)$ but pays $O(n)$ per update;\na plain array updates in $O(1)$ but pays $O(n)$ per range sum. Fenwick and\nsegment trees give us _both_ in $O(\\log n)$. The Fenwick (binary indexed) tree\nis a tiny array keyed by the low bit; the segment tree is a general balanced\ntree over canonical ranges that handles any associative aggregate and, with\nlazy propagation, range updates too.\n",{"path":14140,"title":14141,"module":14113,"summary":14142},"\u002Falgorithms\u002Fdata-structures\u002Fspatial-data-structures","Spatial Data Structures","A balanced BST orders keys on a line, but points in the plane have no single\nnatural order. Quadtrees subdivide space recursively into quadrants; k-d trees\nsplit on alternating coordinates at the median. Both make range and\nnearest-neighbour queries fast by carving the plane into boxes a query can\nprune away. Range trees nest a y-tree in an x-tree for fast orthogonal range\nreporting; interval trees index intervals to answer stabbing queries.\n",{"path":14144,"title":14145,"module":14113,"summary":14146},"\u002Falgorithms\u002Fdata-structures\u002Fskip-lists-and-probabilistic-structures","Skip Lists & Probabilistic Structures","Balanced trees achieve $O(\\log n)$ with rotations and invariants; randomization\ngives the same bound far more simply. A skip list is a layered linked list whose\nexpress lanes are chosen by coin flips, giving expected $O(\\log n)$ search and\ninsert with no rebalancing. A Bloom filter trades exactness for space: a bit\narray and a few hashes answer set membership with no false negatives and a\ntunable false-positive rate, but cannot delete.\n",{"path":14148,"title":14149,"module":14113,"summary":14150},"\u002Falgorithms\u002Fdata-structures\u002Fb-trees","B-Trees","When data lives on disk, the cost that dominates is block transfers, not\ncomparisons — and a binary tree of a billion keys is thirty reads deep. A\nB-tree of minimum degree $t$ is short and wide: $t-1$ to $2t-1$ keys per node,\nall leaves at one depth, height $O(\\log_t n)$. Insertion splits a full node on\nthe way down and pushes its median up; deletion borrows or merges to keep nodes\nfull enough. High fan-out is what minimizes disk I\u002FO.\n",{"path":14152,"title":14153,"module":14113,"summary":14154},"\u002Falgorithms\u002Fdata-structures\u002Fdata-stream-algorithms","Data-Stream Algorithms","Most of this course assumes data sits in fast memory, addressable at will.\nExternal sorting relaxed that to a re-readable disk. The streaming model goes\nfurther: items arrive one at a time, are seen once, and must be discarded, with\nonly sublinear, often polylogarithmic, memory. In exchange, the answers are\napproximate and probabilistic. We set up the model, then meet reservoir\nsampling for a uniform sample of an unknown-length stream and Morris counting\nfor an approximate tally in doubly-logarithmic space.\n",{"path":14156,"title":14157,"module":14113,"summary":14158},"\u002Falgorithms\u002Fdata-structures\u002Fstreaming-sketches","Streaming Sketches","Sampling and counting kept a random subset or a single approximate tally.\nSketches go further: fixed, tiny summaries that answer questions about a\nstream's frequencies. We meet the Count–Min sketch for point frequency\nestimation, Misra–Gries for heavy hitters, and HyperLogLog for distinct\ncounts, each trading a controlled error for space that never grows with the\nstream.\n",{"path":14160,"title":14161,"module":14162,"summary":14163},"\u002Falgorithms\u002Fsequences\u002Ftwo-pointers-and-windows","Two Pointers & Sliding Windows","Sequences & Strings","A family of array idioms that collapse an obvious $O(n^2)$ scan into a single\n$O(n)$ pass by maintaining an invariant as indices move. We meet two pointers\n(converging on a sorted array, and a fast\u002Fslow pair for in-place rewriting)\nand the sliding window (fixed and variable size, amortized $O(n)$). The\ncompanion lesson on prefix sums picks up where the window's positivity\nassumption fails.\n",{"path":14165,"title":14166,"module":14162,"summary":14167},"\u002Falgorithms\u002Fsequences\u002Fprefix-sums","Prefix Sums & Difference Arrays","Prefix sums precompute the running total once so that any range-sum query is a\nsingle subtraction, $P[r{+}1]-P[l]$, in $O(1)$. A hash map of prefix\nfrequencies then counts subarrays summing to $k$ in $O(n)$ — even with negative\nentries, where the sliding window fails. The difference-array dual turns $m$\nrange-adds into $O(m+n)$, and the whole idea lifts to 2-D rectangle sums by\ninclusion–exclusion.\n",{"path":14169,"title":14170,"module":14162,"summary":14171},"\u002Falgorithms\u002Fsequences\u002Fmonotonic-stacks","Monotonic Stacks & Queues","A **monotonic stack** keeps its contents sorted by popping every element that\nwould break the order before each push — turning a family of \"previous\u002Fnext\ngreater (or smaller) element\" questions into a single $O(n)$ scan. We trace\nthe next-greater-element routine push by push and prove its amortized bound,\nfuse two such scans to measure the **largest rectangle in a histogram** in\nlinear time, extend the idea to a **monotonic deque** that streams the\n**sliding-window maximum** in $O(n)$, and use asymmetric tie-breaking to\ncount **subarray minimums** without double-counting duplicates.\n",{"path":14173,"title":14174,"module":14162,"summary":14175},"\u002Falgorithms\u002Fsequences\u002Fbinary-search-on-the-answer","Binary Search on the Answer","Binary search locates the boundary of a **monotone predicate** $p(x)$ in\n$O(\\log(\\text{range}))$ probes; sorted arrays are only one instance. We first\nestablish the half-open `while (lo \u003C hi)` template for $\\textsc{lower\\_bound}$\nand $\\textsc{upper\\_bound}$, then generalize to \"binary search on the answer\":\nwhenever feasibility is monotone in a numeric parameter, we binary search the\nparameter itself, calling a feasibility check at each step.\n",{"path":14177,"title":14178,"module":14162,"summary":14179},"\u002Falgorithms\u002Fsequences\u002Fstring-matching","String Matching: Naive & Rabin–Karp","Given a text $T$ of length $n$ and a pattern $P$ of length $m$, find every\noccurrence of $P$ in $T$. The naive scan costs $O(nm)$ and re-reads text it has\nalready seen. Rabin–Karp fixes the first inefficiency with a **rolling hash**:\neach length-$m$ window is summarized by one number, updated in $O(1)$ per slide,\nverified on a hash match to kill collisions, for expected $O(n+m)$. A companion\nlesson removes the re-reading entirely with KMP and the Z-function.\n",{"path":14181,"title":14182,"module":14162,"summary":14183},"\u002Falgorithms\u002Fsequences\u002Fkmp-and-z-function","String Matching: KMP & the Z-Function","Two linear-time matchers that beat Rabin–Karp's expected bound with a\nworst-case guarantee and no randomness. KMP precomputes a **failure function**\n$\\pi$ so a mismatch slides the pattern by $q-\\pi[q-1]$ and the text pointer\nnever backs up, for $O(n+m)$. The **Z-function** computes the longest\nprefix-match at every position via the Z-box, giving the same bound from a\ndifferent angle; the two encodings of a string's self-overlap convert freely.\n",{"path":14185,"title":14186,"module":14162,"summary":14187},"\u002Falgorithms\u002Fsequences\u002Ftries","Tries & Prefix Trees","A **trie** stores a set of strings in a tree keyed by _characters_, so that\ninsert, search, delete, and prefix-test all run in $O(L)$ time — the length\nof the key, _independent of how many keys are stored_. Shared prefixes are\nstored once, which makes tries the natural structure for autocomplete,\nwildcard dictionaries, board word-search, and — over the alphabet $\\{0,1\\}$\n— the maximum-XOR-pair problem. Radix (Patricia) trees compress the chains.\n",{"path":14189,"title":14190,"module":14162,"summary":14191},"\u002Falgorithms\u002Fsequences\u002Fsuffix-arrays-and-aho-corasick","Suffix Arrays, LCP & Aho–Corasick","A **suffix array** sorts all $n$ suffixes of a string, indexing every substring\nat once; built in $O(n\\log n)$, it locates a pattern by binary search in\n$O(m\\log n)$. Its companion **LCP array** (Kasai's $O(n)$ algorithm) counts\ndistinct substrings and finds the longest repeated substring. **Aho–Corasick**\ngeneralises KMP to a whole dictionary: a trie of patterns plus failure links\nscans the text once in $O(\\text{text} + \\text{matches})$ to report every\noccurrence of every pattern. Manacher's algorithm finds all palindromic\nsubstrings in $O(n)$.\n",{"path":14193,"title":14194,"module":14195,"summary":14196},"\u002Falgorithms\u002Fgraphs\u002Frepresentations-and-traversal","Graph Representations and Traversal","Graphs","A graph captures _relationships_ — who connects to whom. We fix the\nvocabulary, weigh the two standard representations (adjacency list versus\nmatrix), then meet the single search skeleton behind everything that follows:\nWhatever-First-Search, and its breadth-first reading, which finds shortest\npaths by number of edges in $O(V + E)$.\n",{"path":14198,"title":14199,"module":14195,"summary":14200},"\u002Falgorithms\u002Fgraphs\u002Fdepth-first-search","Depth-First Search","Swap BFS's queue for a stack and the search plunges instead of fanning out.\nDepth-first search stamps every vertex with discovery and finish times that\nnest like parentheses, classifies each edge as tree, back, forward, or cross,\nand — through the back edge — decides in one pass whether a graph has a cycle.\nThese timestamps underpin topological sort, strong\nconnectivity, and the rest of this module.\n",{"path":14202,"title":14203,"module":14195,"summary":14204},"\u002Falgorithms\u002Fgraphs\u002Ftopological-sort-and-scc","Topological Sort and Strong Connectivity","Directed acyclic graphs model dependencies: tasks that must precede other\ntasks. A _topological order_ lays such a graph out in a line so every edge\npoints forward, and depth-first finish times yield one almost for free.\nWe then ask the harder question for graphs _with_ cycles: which vertices can\nreach each other? The answer is the strongly connected components, found by a\ntwo-pass DFS.\n",{"path":14206,"title":14207,"module":14195,"summary":14208},"\u002Falgorithms\u002Fgraphs\u002Fminimum-spanning-trees","Minimum Spanning Trees","Given a weighted network, how do we connect everything as cheaply as possible?\nThe answer is a minimum spanning tree, and one lemma — the cut property —\njustifies _every_ correct MST algorithm. We prove the cut and cycle\nproperties by exchange arguments, use them to settle uniqueness, and meet the\noldest MST algorithm, Borůvka's, whose parallel component-merging rounds fall\nstraight out of the cut rule.\n",{"path":14210,"title":14211,"module":14195,"summary":14212},"\u002Falgorithms\u002Fgraphs\u002Fkruskal-and-prim","Kruskal and Prim","The two minimum-spanning-tree algorithms you will actually implement.\nKruskal grows a forest edge by edge, cheapest first, using a union-find\nstructure to reject cycle-closing edges; Prim grows one tree outward from a\nroot with a priority queue, exactly Dijkstra rekeyed by attachment cost. Both\ntraced in full on a nine-town graph, with the edge cases, the bottleneck\nproperty, and where each one wins.\n",{"path":14214,"title":14215,"module":14195,"summary":14216},"\u002Falgorithms\u002Fgraphs\u002Fshortest-paths","Shortest Paths","Finding the cheapest route through a weighted network is one of the most-used\nalgorithms in computing, and a single operation — _relaxation_ — underlies\nevery method. We build the primitive, prove the triangle inequality and\noptimal substructure that make it work, then meet Dijkstra's algorithm: the\ngreedy solution for non-negative weights, traced vertex by vertex, with the\ncut argument that proves each extraction is final.\n",{"path":14218,"title":14219,"module":14195,"summary":14220},"\u002Falgorithms\u002Fgraphs\u002Fall-pairs-and-negative-weights","All-Pairs and Negative Weights","Dijkstra's greedy schedule breaks the moment an edge goes negative. We give it\nup for dynamic programming: Bellman-Ford derived as a DP over edge budgets,\nwith its negative-cycle detector, and Floyd-Warshall computing the distance\nbetween _every_ pair of vertices via a DP over which vertices a path may pass\nthrough. We close with Johnson's algorithm and the arbitrage problems that\nnegative cycles encode.\n",{"path":14222,"title":14223,"module":14195,"summary":14224},"\u002Falgorithms\u002Fgraphs\u002Fnetwork-flow","Network Flow","How much can flow through a network from source to sink? We build flow\nnetworks with capacity and conservation constraints, increase a flow by\npushing along augmenting paths in the residual graph, and see how reverse\nedges let the algorithm undo earlier routing. Ford-Fulkerson and its BFS refinement\nEdmonds-Karp find a maximum flow, traced end to end on a worked network.\n",{"path":14226,"title":14227,"module":14195,"summary":14228},"\u002Falgorithms\u002Fgraphs\u002Fmax-flow-min-cut","Max-Flow Min-Cut and Applications","Why is the flow found when no augmenting path remains actually optimal? The\nanswer is a duality theorem: the maximum flow equals the minimum cut. We prove\nit, read the minimum cut off the final residual graph, then derive bipartite\nmatching and a catalog of modeling reductions from the flow\nabstraction — before touching the modern algorithms that supersede\nEdmonds-Karp.\n",{"path":14230,"title":14231,"module":14195,"summary":14232},"\u002Falgorithms\u002Fgraphs\u002Fbridges-and-articulation-points","Bridges & Articulation Points","A **bridge** is an edge whose removal disconnects the graph; an **articulation\npoint** is a vertex whose removal does. Both are single points of failure in a\nnetwork. A single depth-first search computes discovery times and **low-links**,\nand two local criteria — $low[v] > disc[u]$ for bridges, $low[v] \\ge disc[u]$\nfor cut vertices — find them all in $O(V+E)$.\n",{"path":14234,"title":14235,"module":14195,"summary":14236},"\u002Falgorithms\u002Fgraphs\u002Flowest-common-ancestor","Lowest Common Ancestor & Binary Lifting","Given a rooted tree, the lowest common ancestor of $u$ and $v$ is the deepest\nnode that is an ancestor of both. A naive walk answers one query in $O(h)$;\n**binary lifting** precomputes the $2^k$-th ancestor of every node in\n$O(n\\log n)$, then answers $k$-th-ancestor and LCA queries in $O(\\log n)$ each.\nWe derive both jumps, apply them to tree distance, and compare against the\nEuler-tour + RMQ and Tarjan offline alternatives.\n",{"path":14238,"title":14239,"module":14195,"summary":14240},"\u002Falgorithms\u002Fgraphs\u002Ftwo-sat","2-SAT via Implication Graphs","A boolean formula whose every clause has exactly two literals can be solved in\n_linear_ time — even though its three-literal cousin is NP-complete. The idea\nis to read each clause as a pair of implications, build a directed graph on the\n$2n$ literals, and ask a question we already know how to answer: which literals\nshare a strongly connected component? The formula is satisfiable iff no variable\nlands in the same SCC as its own negation, and the SCCs' topological order\nyields a satisfying assignment for free.\n",{"path":14242,"title":14243,"module":14195,"summary":14244},"\u002Falgorithms\u002Fgraphs\u002Feulerian-tours","Eulerian Tours","An **Eulerian tour** uses every _edge_ of a graph exactly once. We give the\nexact parity and balance conditions under which one exists (even degree\nfor undirected graphs, in-degree equal to out-degree for directed) and Hierholzer's\n$O(E)$ algorithm that constructs one by splicing closed sub-tours. We contrast\nthis sharply with the **Hamiltonian** problem (visit every _vertex_ once),\nwhich is NP-complete: visiting edges is easy, visiting vertices is hard.\n",{"path":14246,"title":14247,"module":14195,"summary":14248},"\u002Falgorithms\u002Fgraphs\u002Fbipartite-matching","Bipartite Matching","Pairing applicants to jobs, students to slots, files to disks: all are\n**maximum bipartite matching**. We solve it combinatorially with **augmenting\npaths** (Kuhn's algorithm, $O(VE)$), speed it up to $O(E\\sqrt V)$ with\n**Hopcroft–Karp**, and uncover the structure behind it — **König's theorem**\n(max matching equals min vertex cover) and **Hall's marriage theorem** (a\nperfect matching exists iff every set has enough neighbors).\n",{"path":14250,"title":14251,"module":14252,"summary":14253},"\u002Falgorithms\u002Fgreedy\u002Fthe-greedy-method","The Greedy Method","Greedy Algorithms","A greedy algorithm builds a solution one locally-best choice at a time and\nnever looks back. We isolate the two properties that make this work — the\ngreedy-choice property and optimal substructure — prove the canonical\nactivity-selection algorithm correct with an exchange argument, watch greedy\nfail on the 0\u002F1 knapsack, and glimpse matroids as the theory\nthat says exactly when the greedy method is optimal.\n",{"path":14255,"title":14256,"module":14252,"summary":14257},"\u002Falgorithms\u002Fgreedy\u002Fscheduling-and-intervals","Scheduling & Interval Partitioning","Three classic scheduling problems all yield to greedy algorithms — and all\nthree turn on a single design decision: which key to sort by. Interval\nscheduling sorts by **finish** time to pack the most compatible jobs;\ninterval partitioning sorts by **start** time and proves the rooms needed\nequal the maximum overlap **depth**; minimizing maximum lateness sorts by\n**deadline** and is justified by an adjacent-swap exchange argument.\n",{"path":14259,"title":14260,"module":14252,"summary":14261},"\u002Falgorithms\u002Fgreedy\u002Fhuffman-codes","Huffman Codes","Huffman coding builds a\nprovably optimal prefix-free binary code by repeatedly merging the two least\nfrequent symbols. We develop prefix-free codes as binary trees, give the\nalgorithm with a priority queue, build a Huffman tree from example\nfrequencies, prove optimality with the same greedy-choice-plus-substructure\nargument, and pin the running time at $O(n\\log n)$.\n",{"path":14263,"title":14264,"module":14252,"summary":14265},"\u002Falgorithms\u002Fgreedy\u002Fmatroids","Matroids & Exchange Arguments","The capstone of the greedy module: _why_ and _when_ a greedy algorithm is\nprovably optimal. We recap the two correctness templates — **greedy-stays-ahead**\nand the **exchange argument** — then meet the **matroid** $M=(S,\\mathcal{I})$, an\nabstraction whose **exchange property** is the structure greedy needs.\nThe matroid–greedy theorem says sorting by weight and taking what stays\nindependent yields a maximum-weight basis _if and only if_ the structure is a\nmatroid. Kruskal's MST is the canonical instance; 0\u002F1 knapsack and TSP are the\ncanonical failures.\n",{"path":14267,"title":14268,"module":14252,"summary":14269},"\u002Falgorithms\u002Fgreedy\u002Fstable-matching","Stable Matching (Gale–Shapley)","Two sides each rank the other; we want a matching with no **blocking pair** — no\ntwo participants who both prefer each other to their assigned partners. The\n**Gale–Shapley deferred-acceptance** algorithm has proposers propose in\npreference order while receivers tentatively hold the best offer so far. We prove\nit terminates in $\\O(n^2)$ proposals, returns a **perfect** matching, and that\nthe matching is **stable**. A sharper asymmetry follows: deferred acceptance is\n**proposer-optimal** and **receiver-pessimal**, the structural fact behind the\nresidency match and school-choice systems.\n",{"path":14271,"title":14272,"module":14273,"summary":14274},"\u002Falgorithms\u002Fdynamic-programming\u002Fprinciples","Principles of Dynamic Programming","Dynamic Programming","Dynamic programming is recursion with memory: when a recursive solution\nre-solves the same subproblems again and again, we solve each one once and\nstore the answer. We identify the two structural conditions that make this\nwork — overlapping subproblems and optimal substructure — contrast top-down\nmemoization with bottom-up tabulation, and distil the whole method into a\nfive-step recipe.\n",{"path":14276,"title":14277,"module":14273,"summary":14278},"\u002Falgorithms\u002Fdynamic-programming\u002Fsequence-dp","Sequence Alignment & LCS","Two strings can be compared by how much of one appears inside the\nother. The longest common subsequence (LCS) and edit distance are the two\nclassic measures, and they are the _same_ dynamic program with different\ncosts. We derive the LCS recurrence by examining the last characters, fill a\nworked DP table, reconstruct the subsequence, and then show edit distance as\nthe identical $\\Theta(mn)$ pattern.\n",{"path":14280,"title":14281,"module":14273,"summary":14282},"\u002Falgorithms\u002Fdynamic-programming\u002Flongest-increasing-subsequence","Longest Increasing Subsequence","Given a sequence of numbers, how long is its longest strictly increasing\nsubsequence? A first dynamic program indexes subproblems by the element each\nsubsequence _ends at_, giving an $O(n^2)$ solution with parent-pointer\nreconstruction. A sharper idea, the patience-sorting _tails_ array searched by\nbinary search, drops the time to $O(n\\log n)$. We then fold in the\nvariants: non-decreasing, counting, Russian-doll envelopes, and bitonic.\n",{"path":14284,"title":14285,"module":14273,"summary":14286},"\u002Falgorithms\u002Fdynamic-programming\u002Fknapsack","Knapsack & Subset Problems","We start from $\\textsc{Subset-sum}$ — does some sublist hit a target $t$? — and its\ninclude\u002Fexclude recurrence over a boolean table $A(i, u)$, then bolt on values\nto get 0\u002F1 knapsack as the same machine with $\\lor$ promoted to $\\max$. We fill\nboth tables, recover the chosen items, and confront the surprise that the\n$\\Theta(nt)$ running time is only _pseudo-polynomial_ — exponential in the bit\nlength $b$, and unimprovable unless $\\mathrm{P}=\\mathrm{NP}$ since subset-sum is\n$\\textsc{NP-complete}$. The fractional variant reveals the sharp line between greedy\nand dynamic programming.\n",{"path":14288,"title":14289,"module":14273,"summary":14290},"\u002Falgorithms\u002Fdynamic-programming\u002Fcoin-change-and-unbounded","Coin Change & Unbounded Knapsack","The previous lesson let each item be taken at most once. Drop that cap — items\nmay be reused _any number of times_ — and the 0\u002F1 knapsack collapses from a\ntwo-dimensional table to a one-dimensional one, because there is no longer a\nprefix of \"already-used\" items to track. We meet **unbounded knapsack**, then\nits most famous instance, **coin change**: the minimum-coins recurrence\n$C[a] = 1 + \\min_c C[a-c]$, and the counting variant where the _order of the\nloops_ decides whether you count unordered combinations or ordered sequences —\nthe classic bug. Greed fails in general but works for canonical coin systems.\n",{"path":14292,"title":14293,"module":14273,"summary":14294},"\u002Falgorithms\u002Fdynamic-programming\u002Finterval-dp","Interval DP","Many problems ask for the best way to combine a contiguous range of items, and\nthe answer is a dynamic program over subintervals $[i,j]$ that chooses a split\npoint $k$. We derive the pattern from matrix-chain multiplication —\nparenthesising a product to minimize scalar multiplications in $O(n^3)$ — distil\nit into a reusable template filled by increasing interval length, and then meet\nits sharpest variant: the \"last operation\" trick behind Burst Balloons and\ncutting a stick, where fixing the _last_ move (not the first) makes the two\nsides independent.\n",{"path":14296,"title":14297,"module":14273,"summary":14298},"\u002Falgorithms\u002Fdynamic-programming\u002Ftree-dp","Dynamic Programming on Trees","When the subproblems of a dynamic program are _rooted subtrees_, a single\npost-order DFS solves the whole thing in $O(n)$: each node combines the\nalready-computed answers of its children. We meet the archetype — maximum-weight\nindependent set on a tree — then the \"path through a node\" pattern behind tree\ndiameter and maximum path sum, and finally **rerooting**, which computes a\nper-node answer for _every_ node as root in $O(n)$ with two passes.\n",{"path":14300,"title":14301,"module":14273,"summary":14302},"\u002Falgorithms\u002Fdynamic-programming\u002Fbitmask-dp","Bitmask DP","When a subproblem depends not on an index or a prefix but on _which subset_ of\na small ground set has been used, we can encode that subset as the bits of an\ninteger and index a DP table by it. With $n \\le \\sim 20$ the $2^n$ subsets fit\nin a table, turning $\\Theta(n!)$ brute force into $O(2^n \\cdot \\text{poly}(n))$.\nWe meet the bit tricks, the Held–Karp TSP archetype, assignment by mask,\nsubset-sum partitioning, and submask enumeration with its $3^n$ bound.\n",{"path":14304,"title":14305,"module":14273,"summary":14306},"\u002Falgorithms\u002Fdynamic-programming\u002Fdp-optimizations","DP Optimizations","A correct DP recurrence is only half the battle; its naive evaluation is often\na factor of $n$ slower than necessary. This capstone surveys five techniques,\nmonotonic-queue, the convex hull trick, divide-and-conquer optimization,\nKnuth's optimization, and SOS DP, that each exploit _structure in the\ntransition_ (a sliding window, linear costs, monotone optimal splits, the\nquadrangle inequality, or subset lattices) to shave an $O(n)$, $O(\\log n)$, or\nworse factor off the running time.\n",{"path":14308,"title":14309,"module":14273,"summary":14310},"\u002Falgorithms\u002Fdynamic-programming\u002Fdp-on-graphs","Dynamic Programming on Graphs","Many graph algorithms are dynamic programs: the subproblem is the\n_best value reachable under a restricted resource_ — intermediate vertices\nallowed, edges allowed, or a topological prefix — and edge _relaxation_ is the\nDP transition. We frame Floyd–Warshall as the archetype ($O(V^3)$ all-pairs\nshortest paths), Bellman–Ford as a DP over path length (the at-most-$K$-stops\nvariant), DAG-DP in topological order ($O(V+E)$), and Warshall's transitive\nclosure as the boolean analog.\n",{"path":14312,"title":14313,"module":14273,"summary":14314},"\u002Falgorithms\u002Fdynamic-programming\u002Fdigit-and-probability-dp","Digit & Probability DP","Two DP patterns with unusual state. _Digit DP_ counts the\nintegers in a range $[L, R]$ that satisfy a digit constraint by walking the\ndecimal places of the bound, carrying a _tight_ flag that marks when the prefix\nstill equals the bound's. _Probability\u002FExpectation DP_ replaces \"best value\" with\n\"expected value,\" using linearity of expectation to make each state an\naverage over its weighted transitions — the natural tool for expected step\ncounts and absorbing Markov chains.\n",{"path":14316,"title":14317,"module":14318,"summary":14319},"\u002Falgorithms\u002Fbacktracking\u002Fbacktracking-fundamentals","Backtracking: Subsets, Permutations & Combinations","Backtracking & Search","Backtracking builds a solution one choice at a time and abandons a partial\nsolution the moment it cannot be completed, exploring a state-space tree by\ndepth-first search. We meet the universal choose\u002Fexplore\u002Fun-choose template,\nderive the canonical enumerations — subsets ($2^n$), permutations ($n!$), and\ncombinations ($\\binom{n}{k}$) — handle duplicate elements by skipping equal\nsiblings, and see how pruning turns an exponential search into a tractable one.\n",{"path":14321,"title":14322,"module":14318,"summary":14323},"\u002Falgorithms\u002Fbacktracking\u002Fconstraint-search","Constraint Search: N-Queens & Sudoku","Many hard puzzles are **constraint satisfaction problems**: assign each\nvariable a value from its domain so that every constraint holds. Backtracking\nsolves them by assigning variables one at a time and rejecting a partial\nassignment the instant a constraint breaks. We make the rejection cheap — $O(1)$\nconflict checks for N-Queens via column and diagonal sets — and prune harder\nwith **forward checking**, **MRV** ordering, and **constraint propagation**,\nwhich is what lets an exponential search actually finish.\n",{"path":14325,"title":14326,"module":14318,"summary":14327},"\u002Falgorithms\u002Fbacktracking\u002Fbranch-and-bound","Branch & Bound and Meet in the Middle","Plain backtracking prunes a search tree by _feasibility_; for _optimization_\nproblems we can prune far more aggressively by _value_. **Branch and bound**\nkeeps the best complete solution found so far and discards any partial solution\nwhose optimistic bound cannot beat it. **Meet in the middle** splits the\ninstance in two, enumerates each half, and recombines by binary search — turning\n$2^n$ into $O(2^{n\u002F2}\\,n)$ and pushing exact search out to $n \\approx 40$.\n",{"path":14329,"title":14330,"module":14318,"summary":14331},"\u002Falgorithms\u002Fbacktracking\u002Fgraph-backtracking","Graph Backtracking: m-Coloring & Hamiltonian Paths","Two famous graph problems have no known efficient algorithm, yet yield cleanly\nto backtracking with the right pruning. **Graph $m$-coloring** assigns one of\n$m$ colors to each vertex so no edge is monochromatic; we color vertices in turn\nand reject a color the instant a neighbor already has it. **Hamiltonian\npath\u002Fcycle** asks for a walk visiting every vertex exactly once; we extend a path\ngreedily and backtrack on dead ends. Both are NP-complete, so the worst case is\nexponential — but feasibility pruning and good vertex ordering make real\ninstances tractable, and the contrast with the easy Eulerian condition shows why.\n",{"path":14333,"title":14334,"module":14335,"summary":14336},"\u002Falgorithms\u002Fmathematical-algorithms\u002Fnumber-theory-basics","Number Theory: GCD & Modular Arithmetic","Mathematical Algorithms","This lesson opens the mathematical-algorithms module with the bedrock of\ncomputational number theory. We prove Euclid's recurrence\n$\\gcd(a,b)=\\gcd(b,\\,a\\bmod b)$ and its $O(\\log\\min(a,b))$ running time, extend\nit to recover Bézout coefficients $x,y$ with $ax+by=\\gcd(a,b)$, and build\nmodular arithmetic on residue classes — including when a modular inverse\n$a^{-1}\\bmod m$ exists and how to compute it.\n",{"path":14338,"title":14339,"module":14335,"summary":14340},"\u002Falgorithms\u002Fmathematical-algorithms\u002Fmodular-exponentiation-and-primality","Modular Exponentiation & Primality","Computing $a^n \\bmod m$ naively costs $n$ multiplications; **repeated squaring**\ndoes it in $O(\\log n)$ by reading the bits of the exponent. We use this routine\nto state **Fermat's little theorem** (and the modular inverse it gives), then to\ntest primality — trial division, the probabilistic **Fermat** and **Miller–Rabin**\ntests, and the deterministic witness set that settles primality for every 64-bit\nnumber.\n",{"path":14342,"title":14343,"module":14335,"summary":14344},"\u002Falgorithms\u002Fmathematical-algorithms\u002Fsieve-and-factorization","Sieves & Factorization","The previous lesson tested one number for primality; here we ask for _all_\nprimes up to $n$ at once. The **sieve of Eratosthenes** cross-cuts composites\nin $O(n\\log\\log n)$, and a **linear sieve** does it in $O(n)$ while recording\neach number's **smallest prime factor**, which then factors any $x \\le n$ in\n$O(\\log x)$. From a factorization $x = \\prod p_i^{e_i}$ the multiplicative\nfunctions $\\tau$, $\\sigma$, and Euler's totient $\\varphi$ fall out immediately.\n",{"path":14346,"title":14347,"module":14335,"summary":14348},"\u002Falgorithms\u002Fmathematical-algorithms\u002Fcombinatorics","Combinatorics & Counting","Counting is the arithmetic of finite sets. We build up from permutations\n$n!$ and combinations $\\binom{n}{k}$, prove Pascal's rule by a bijection,\nand count multisets with stars and bars. The practical core is computing\n$\\binom{n}{k}\\bmod p$ in $O(1)$ from precomputed factorials and inverse\nfactorials. We close with inclusion–exclusion and the Chinese Remainder\nTheorem, both of which lean on the modular inverse from the previous lesson.\n",{"path":14350,"title":14351,"module":14335,"summary":14352},"\u002Falgorithms\u002Fmathematical-algorithms\u002Fmatrix-exponentiation","Matrix Exponentiation","A linear recurrence advances by a fixed linear rule, so one step is a\n**matrix–vector** product and $n$ steps are a **matrix power**. Packaging\nFibonacci, and any $k$-term recurrence, into a transition matrix lets us jump\nto the $n$-th term in $O(k^3 \\log n)$ by **exponentiation by squaring** — the\nsame doubling trick from modular exponentiation, now over matrices.\n",{"path":14354,"title":14355,"module":14335,"summary":14356},"\u002Falgorithms\u002Fmathematical-algorithms\u002Ffast-fourier-transform","Fast Fourier Transform","Multiplying two degree-$n$ polynomials by the schoolbook method costs\n$\\Theta(n^2)$. Evaluating them at the **$n$-th roots of unity** turns\nmultiplication into pointwise products, and the **Cooley–Tukey FFT** computes\nall those evaluations in $\\Theta(n\\log n)$ by splitting even and odd\ncoefficients. The inverse FFT interpolates back, giving $\\Theta(n\\log n)$\npolynomial and big-integer multiplication.\n",{"path":14358,"title":14359,"module":14335,"summary":14360},"\u002Falgorithms\u002Fmathematical-algorithms\u002Fgradient-descent","Numerical Optimization and Gradient Descent","Most of this course chases **discrete** optima over finite structures; here the\nsearch space is **continuous** and the objective $f$ is differentiable. The\n**gradient** points uphill, so stepping against it —\n$x_{t+1} = x_t - \\eta\\,\\nabla f(x_t)$ — walks downhill. **Convexity** makes every\nlocal minimum global; for convex $L$-smooth $f$ gradient descent converges at\n$O(1\u002Ft)$, and **geometrically** under strong convexity. **Newton's method** uses\nthe Hessian for local quadratic convergence, and **bisection** is the robust\nbracketing fallback for roots.\n",{"path":14362,"title":14363,"module":14364,"summary":14365},"\u002Falgorithms\u002Fcomputational-geometry\u002Fgeometric-primitives","Geometric Primitives & Orientation","Computational Geometry","Computational geometry is built on a single reliable primitive — the\n**orientation test**, a sign of a cross product that tells whether three points\nturn left, right, or lie collinear. From points-as-vectors and the dot and\ncross products we derive orientation, segment intersection, the shoelace area\nformula, and point-in-polygon tests, keeping all arithmetic **exact and\ninteger** so that no floating-point rounding can corrupt a sign.\n",{"path":14367,"title":14368,"module":14364,"summary":14369},"\u002Falgorithms\u002Fcomputational-geometry\u002Fconvex-hull","Convex Hull","The convex hull is the smallest convex polygon enclosing a point set — the\nrubber band snapped around the nails. We build it with Andrew's monotone chain,\nsorting by $(x,y)$ and sweeping a lower and upper hull while popping any\nnon-left turn via the orientation primitive, in $O(n\\log n)$. A reduction from\nsorting shows that bound is optimal, and the hull yields diameter, smallest\nenclosing rectangle, and more through rotating calipers.\n",{"path":14371,"title":14372,"module":14364,"summary":14373},"\u002Falgorithms\u002Fcomputational-geometry\u002Fsweep-line","Sweep-Line Algorithms","The plane-sweep paradigm turns a static $2$-D geometry problem into a dynamic\n$1$-D ordered-set problem: a vertical line sweeps left to right, stopping at an\n$x$-sorted **event queue** while a balanced-BST **status structure** tracks the\nobjects it currently crosses, ordered by $y$. We derive Bentley–Ottmann segment\nintersection in $O((n+k)\\log n)$, recover closest-pair in $O(n\\log n)$, and\nreduce skyline, rectangle-area, and overlap problems to $\\pm1$ event sweeps.\n",{"path":14375,"title":14376,"module":14364,"summary":14377},"\u002Falgorithms\u002Fcomputational-geometry\u002Fpolygons-and-proximity","Polygons & Proximity","Four classics that live on top of the orientation primitive and the convex\nhull. **Closest pair** falls to divide-and-conquer in $\\Theta(n\\log n)$, where a\npacking argument caps the cross-boundary combine at seven neighbours per point.\n**Point-in-polygon** is the ray-casting parity test or the winding-number count\nthat also handles self-intersecting boundaries, both with their edge caveats. The **shoelace formula**\ngives signed area as a sum of cross products, and **rotating calipers** walk the\nhull to read off diameter and width in $O(n)$.\n",{"path":14379,"title":14380,"module":14381,"summary":14382},"\u002Falgorithms\u002Fintractability\u002Fp-np-reductions","P, NP, and Reductions","Intractability","Most problems we have met so far have fast algorithms. A vast and important\nfamily seemingly does not. This lesson builds the vocabulary for that\ndivide: decision problems, the class $\\mathsf{P}$ of problems we can solve\nquickly, the class $\\mathsf{NP}$ of problems whose solutions we can _check_\nquickly, and polynomial-time reductions, the tool that lets us compare the\ndifficulty of two problems without solving either.\n",{"path":14384,"title":14385,"module":14381,"summary":14386},"\u002Falgorithms\u002Fintractability\u002Fnp-completeness","NP-Completeness","Some problems in $\\mathsf{NP}$ are universally hardest: every other problem\nin $\\mathsf{NP}$ reduces to them. This lesson defines $\\mathsf{NP}$-hard and\n$\\mathsf{NP}$-complete, states the Cook–Levin theorem that anchors the\ntheory on **SAT**, walks the web of reductions that grows from it, and gives\nthe four-step recipe for proving a brand-new problem $\\mathsf{NP}$-complete.\n",{"path":14388,"title":14389,"module":14381,"summary":14390},"\u002Falgorithms\u002Fintractability\u002Fcoping-with-hardness","Coping with NP-Hardness","An $\\mathsf{NP}$-hardness proof rules out an exact polynomial-time algorithm,\nnot the need for answers. This lesson surveys four practical responses to\nhardness: approximation algorithms with a provable ratio (worked through a\n2-approximation for vertex cover), heuristics and local search, exact\nexponential methods like branch and bound, and exploiting special structure\nin the instances you actually face.\n",{"path":14392,"title":14393,"module":14381,"summary":14394},"\u002Falgorithms\u002Fintractability\u002Fapproximation-algorithms","Approximation Algorithms","When a problem is $\\mathsf{NP}$-hard we can still ask for a solution\nprovably close to optimal. This lesson makes the approximation ratio\n$\\rho$ precise, separates absolute from relative guarantees, and proves the\nratios of four classic algorithms: greedy set cover ($H_n \\approx \\ln n$),\nthe MST-doubling $2$-approximation for metric TSP, load balancing, and the\nknapsack FPTAS. It closes with the hierarchy PTAS \u002F FPTAS and the limits of\ninapproximability.\n",{"path":14396,"title":14397,"module":6,"summary":6},"\u002Falgorithms","Algorithms",{"path":14399,"title":14400,"module":14401,"summary":14402},"\u002Fcalculus\u002Flimits-and-continuity\u002Ffunctions-and-models","Functions and Mathematical Models","Limits and Continuity","A function assigns exactly one output to each input and can be presented four ways: verbally, numerically, graphically, or by a formula. The elementary families — linear, polynomial, power, rational, trigonometric, exponential — model most elementary phenomena, and transformation, combination, and composition build every other function from them.\n",{"path":14404,"title":14405,"module":14401,"summary":14406},"\u002Fcalculus\u002Flimits-and-continuity\u002Fthe-limit-of-a-function","The Limit of a Function","The tangent and velocity problems both ask for a value a ratio approaches but never reaches — the limit. Its intuitive two-sided form splits into one-sided limits that must agree; a limit fails to exist when they disagree or when the function grows without bound, the latter producing a vertical asymptote.\n",{"path":14408,"title":14409,"module":14401,"summary":14410},"\u002Fcalculus\u002Flimits-and-continuity\u002Flimit-laws-and-the-precise-definition","Limit Laws and the ε–δ Definition","The Limit Laws reduce a limit to arithmetic on simpler limits, and direct substitution settles polynomials and rational functions outright. The 0\u002F0 forms that resist substitution yield to algebra or the Squeeze Theorem, and the ε–δ definition makes \"arbitrarily close\" precise as a pair of quantified inequalities.\n",{"path":14412,"title":14413,"module":14401,"summary":14414},"\u002Fcalculus\u002Flimits-and-continuity\u002Fcontinuity","Continuity","A function is continuous at a point when its limit there equals its value, so the graph has no break. Continuity fails in three geometric ways; it is closed under arithmetic and composition, so the elementary families and their combinations are continuous; and on a closed interval it forces the Intermediate Value Theorem, which locates roots.\n",{"path":14416,"title":14417,"module":14418,"summary":14419},"\u002Fcalculus\u002Fderivatives\u002Fthe-derivative-and-rates-of-change","The Derivative and Rates of Change","Derivatives","A single limit with three readings: the slope of the tangent line, the instantaneous velocity of a moving object, and the rate of change of one quantity with respect to another. Built from the difference quotient, extended from a value at one point to a function of x, and undefined exactly where a corner, jump, or vertical tangent appears.\n",{"path":14421,"title":14422,"module":14418,"summary":14423},"\u002Fcalculus\u002Fderivatives\u002Fdifferentiation-rules-and-the-chain-rule","Differentiation Rules and the Chain Rule","Computing every derivative from the limit definition is tedious. A short list of rules — power, constant multiple, sum, product, quotient — differentiates any polynomial or rational function by inspection. The trigonometric derivatives follow from one limit, and the chain rule extends everything to composite functions by multiplying rates along the composition.\n",{"path":14425,"title":14426,"module":14418,"summary":14427},"\u002Fcalculus\u002Fderivatives\u002Fimplicit-differentiation-and-related-rates","Implicit Differentiation and Related Rates","Not every curve is the graph of y = f(x). Implicit differentiation finds a slope from an equation in x and y directly, treating y as an unknown function and differentiating both sides. The same chain-rule idea drives related rates, where one measured rate of change forces another through a geometric constraint, and interprets the derivative as a rate across the sciences.\n",{"path":14429,"title":14430,"module":14418,"summary":14431},"\u002Fcalculus\u002Fderivatives\u002Flinear-approximations-and-differentials","Linear Approximations and Differentials","A differentiable curve looks like its tangent line under enough magnification, so the tangent is a usable stand-in for the function near the point of contact. The linear approximation and its linearization, written in the language of differentials dy and dx, estimate both function values and the measurement error propagated into a computed quantity.\n",{"path":14433,"title":14434,"module":14435,"summary":14436},"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fextrema-and-the-mean-value-theorem","Extrema and the Mean Value Theorem","Applications of Derivatives","Absolute and local extrema, the Extreme Value Theorem that guarantees them, and Fermat's Theorem pinning candidates to critical numbers. The Closed Interval Method turns the search for extrema into a finite checklist. Rolle's Theorem and the Mean Value Theorem then connect a function's values to its derivative, giving the tool that most of differential calculus rests on.\n",{"path":14438,"title":14439,"module":14435,"summary":14440},"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fhow-derivatives-shape-a-graph","How Derivatives Shape a Graph","The sign of the first derivative fixes where a function rises and falls, and a sign change identifies each local extremum through the First Derivative Test. The second derivative sets concavity and inflection points and gives a faster Second Derivative Test. Limits at infinity describe end behavior and the horizontal asymptotes a curve settles toward.\n",{"path":14442,"title":14443,"module":14435,"summary":14444},"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fcurve-sketching-and-optimization","Curve Sketching and Optimization","A checklist that synthesizes domain, symmetry, asymptotes, monotonicity, extrema, and concavity into a hand sketch of any function, plus the slant asymptote for rational functions whose degree exceeds the denominator's. The same extremum machinery, applied to a word problem, becomes the optimization template: model one quantity, reduce it to a function of a single variable, and find its absolute extremum.\n",{"path":14446,"title":14447,"module":14435,"summary":14448},"\u002Fcalculus\u002Fapplications-of-derivatives\u002Fnewtons-method-and-antiderivatives","Newton's Method and Antiderivatives","Newton's method solves $f(x) = 0$ by repeatedly replacing the curve with its tangent line and jumping to the tangent's root, converging fast when it works and diverging when the derivative is small. Antiderivatives reverse differentiation: every antiderivative of a function differs from another by a constant, so the general antiderivative is a family of parallel curves, pinned to one by an initial condition.\n",{"path":14450,"title":14451,"module":14452,"summary":14453},"\u002Fcalculus\u002Fintegrals\u002Farea-and-the-definite-integral","Area and the Definite Integral","Integrals","The area under a curve is defined as a limit of sums of rectangle areas. The same limit — a Riemann sum taken as the mesh shrinks to zero — defines the definite integral, a single number measuring signed area, total distance, and every accumulated quantity built the same way. Its properties, comparison bounds, and reading as net area follow directly from the limit.\n",{"path":14455,"title":14456,"module":14452,"summary":14457},"\u002Fcalculus\u002Fintegrals\u002Fthe-fundamental-theorem-of-calculus","The Fundamental Theorem of Calculus","Differentiation and integration are inverse operations. Part 1 says the derivative of an area-accumulation function is the integrand; Part 2 says a definite integral equals the change in any antiderivative across the interval. Together they replace limits of Riemann sums with antiderivative lookups, define the indefinite integral, and give the Net Change Theorem for rates.\n",{"path":14459,"title":14460,"module":14452,"summary":14461},"\u002Fcalculus\u002Fintegrals\u002Fthe-substitution-rule","The Substitution Rule","Substitution runs the Chain Rule backward: spotting an inner function whose derivative also appears in the integrand lets the variable change to $u$ and collapse a composite integral to a simple one. The rule applies to indefinite and definite integrals, with two ways to handle the limits, and it yields the symmetry shortcuts that double even integrands and vanish odd ones.\n",{"path":14463,"title":14464,"module":14465,"summary":14466},"\u002Fcalculus\u002Fapplications-of-integration\u002Fareas-and-volumes","Areas Between Curves and Volumes","Applications of Integration","A definite integral computes any quantity that a limit of Riemann sums approximates. Applied to geometry it gives the area between two curves and the volume of a solid: by cross-sections, by disks and washers when the region is revolved, and by cylindrical shells when inverting the boundary is awkward.\n",{"path":14468,"title":14469,"module":14465,"summary":14470},"\u002Fcalculus\u002Fapplications-of-integration\u002Fwork-average-value-and-arc-length","Work, Average Value, Arc Length, and Surface Area","The work done by a force that varies with position, the average value of a function and the Mean Value Theorem it satisfies, the length of a curve, and the area of a surface swept out by revolving that curve. Each is a limit of Riemann sums, hence a definite integral.\n",{"path":14472,"title":14473,"module":14465,"summary":14474},"\u002Fcalculus\u002Fapplications-of-integration\u002Fphysics-economics-and-probability","Applications to Physics, Economics, and Probability","Definite integrals in physics, economics, and statistics: the force a fluid exerts on a submerged plate, the balance point of a plane region, the money consumers save at a market price, and the probability that a continuous random variable lands in an interval, together with its mean.\n",{"path":14476,"title":14477,"module":14478,"summary":14479},"\u002Fcalculus\u002Fexponential-logarithmic-and-inverse-functions\u002Finverse-functions-logarithms-and-exponentials","Inverse Functions, Logarithms, and Exponentials","Exponential, Logarithmic, and Inverse Functions","A one-to-one function has an inverse that reverses it, with a graph mirrored across y = x and a derivative given by the reciprocal-slope rule. The exponential e^x is its own derivative and the natural logarithm has derivative 1\u002Fx; logarithmic differentiation turns products, quotients, and variable powers into sums.\n",{"path":14481,"title":14482,"module":14478,"summary":14483},"\u002Fcalculus\u002Fexponential-logarithmic-and-inverse-functions\u002Fgrowth-decay-inverse-trig-and-hyperbolic-functions","Growth, Decay, Inverse Trigonometric, and Hyperbolic Functions","Any quantity whose rate of change is proportional to its size grows or decays exponentially, the single equation y' = ky behind populations, radioactive decay, cooling, and continuously compounded interest. The inverse trigonometric functions have algebraic derivatives, and the hyperbolic functions, built from e^x and e^{-x}, describe the hanging cable.\n",{"path":14485,"title":14486,"module":14478,"summary":14487},"\u002Fcalculus\u002Fexponential-logarithmic-and-inverse-functions\u002Flhospitals-rule","Indeterminate Forms and l'Hospital's Rule","When a limit produces 0\u002F0 or infinity over infinity, the value is undetermined by the forms alone. l'Hospital's Rule resolves both by replacing the ratio of functions with the ratio of their derivatives. Products, differences, and powers reduce to a quotient the rule can handle, and repeated use ranks the growth of logarithms, powers, and exponentials.\n",{"path":14489,"title":14490,"module":14491,"summary":14492},"\u002Fcalculus\u002Ftechniques-of-integration\u002Fintegration-by-parts","Integration by Parts","Techniques of Integration","The product rule for derivatives reverses into integration by parts, trading the integral of $u\\,\\d v$ for the integral of $v\\,\\d u$ whenever the second is easier. The LIATE ordering fixes which factor to differentiate. Standard cases: a polynomial against a transcendental factor, repeated parts, cyclic integrals that solve for themselves, and reduction formulas that peel an exponent down by recursion.\n",{"path":14494,"title":14495,"module":14491,"summary":14496},"\u002Fcalculus\u002Ftechniques-of-integration\u002Ftrigonometric-integrals-and-substitution","Trigonometric Integrals and Substitution","Two related techniques. Trigonometric integrals evaluate powers and products of sine, cosine, tangent, and secant by splitting off one factor and converting the rest with a Pythagorean identity, or by dropping even powers with half-angle formulas. Trigonometric substitution runs the idea in reverse: replace x by a sine, tangent, or secant to clear a radical, integrate, then read the answer back off a reference triangle.\n",{"path":14498,"title":14499,"module":14491,"summary":14500},"\u002Fcalculus\u002Ftechniques-of-integration\u002Fpartial-fractions-and-integration-strategy","Partial Fractions and Integration Strategy","Any rational function integrates in closed form: factor the denominator, split the fraction into simple pieces by partial fractions, and integrate each piece as a logarithm or an arctangent. Four denominator cases exhaust the possibilities. A four-step strategy then sorts an arbitrary integrand by its shape to the technique that fits it, and a short catalog records elementary functions whose antiderivatives are not elementary.\n",{"path":14502,"title":14503,"module":14491,"summary":14504},"\u002Fcalculus\u002Ftechniques-of-integration\u002Fapproximate-and-improper-integrals","Approximate and Improper Integrals","Two definite integrals the Fundamental Theorem cannot reach. With no antiderivative available, the Midpoint, Trapezoidal, and Simpson rules approximate the integral from sample values, each carrying a provable error bound. With an infinite interval or an integrand that blows up, the improper integral is defined as a limit that either converges or diverges; the Comparison Test settles which without evaluating it.\n",{"path":14506,"title":14507,"module":14508,"summary":14509},"\u002Fcalculus\u002Fparametric-and-polar\u002Fparametric-curves-and-their-calculus","Parametric Curves and Their Calculus","Parametric Equations and Polar Coordinates","A parametric curve gives x and y separately as functions of a third variable, recording not only a path but the direction and timing with which it is traced. Eliminating the parameter recovers a Cartesian equation; the slope, area, arc-length, and surface-area formulas run directly on the parameter, with the cycloid and astroid as worked examples.\n",{"path":14511,"title":14512,"module":14508,"summary":14513},"\u002Fcalculus\u002Fparametric-and-polar\u002Fpolar-coordinates","Polar Coordinates","Polar coordinates locate a point by a distance from the pole and an angle from the polar axis, giving circles, spirals, and flower-shaped curves short equations. Conversion between the two systems is right-triangle trigonometry, and treating a polar curve as a parametric curve in the angle yields the tangent, area, and arc-length formulas.\n",{"path":14515,"title":14516,"module":14508,"summary":14517},"\u002Fcalculus\u002Fparametric-and-polar\u002Fconic-sections","Conic Sections","Parabolas, ellipses, and hyperbolas are the plane curves cut from a double cone. Each has a focus-based geometric definition and a standard Cartesian equation. A single number, the eccentricity, ties the three together, and placing a focus at the pole gives all of them one polar equation that describes planetary orbits.\n",{"path":14519,"title":13116,"module":14520,"summary":14521},"\u002Fcalculus\u002Fsequences-and-series\u002Fsequences","Infinite Sequences and Series","A sequence is a function on the positive integers, and its limit is defined almost exactly like a limit at infinity. The Limit Laws and Squeeze Theorem carry over from functions, monotonic and bounded sequences give a convergence criterion, and the Monotonic Sequence Theorem guarantees a limit exists without naming it.\n",{"path":14523,"title":14524,"module":14520,"summary":14525},"\u002Fcalculus\u002Fsequences-and-series\u002Fseries-and-the-integral-test","Series and the Integral Test","Adding infinitely many terms is made precise as the limit of partial sums. The two series with closed-form partial sums are geometric and telescoping; the harmonic series diverges even as its terms shrink to zero. The Integral Test compares a positive series to an improper integral, settling the p-series and supplying a remainder bound for estimating sums.\n",{"path":14527,"title":14528,"module":14520,"summary":14529},"\u002Fcalculus\u002Fsequences-and-series\u002Fthe-convergence-tests","The Convergence Tests","The comparison, alternating-series, ratio, and root tests decide convergence without a closed-form partial sum. Absolute convergence is stronger than conditional convergence and is preserved under rearrangement; an alternating series errs by less than its first omitted term. A test is chosen from the shape of the general term.\n",{"path":14531,"title":14532,"module":14520,"summary":14533},"\u002Fcalculus\u002Fsequences-and-series\u002Fpower-series","Power Series","A power series is a polynomial of infinite degree whose convergence set is an interval centered at $a$, with a radius the Ratio Test finds and endpoints that must be tested by hand. Inside that interval the series represents a function that can be differentiated and integrated term by term, generating new representations from the geometric series.\n",{"path":14535,"title":14536,"module":14520,"summary":14537},"\u002Fcalculus\u002Fsequences-and-series\u002Ftaylor-and-maclaurin-series","Taylor and Maclaurin Series","If a function equals a power series, its coefficients are forced: the nth is the nth derivative at the center over n factorial. We derive that formula, use Taylor's Inequality to prove the standard series for the exponential, sine, and cosine, record the binomial series and a reference table, and bound the error when a Taylor polynomial replaces a function.\n",{"path":14539,"title":14540,"module":14541,"summary":14542},"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fvectors-and-the-dot-product","Three-Dimensional Coordinates, Vectors, and the Dot Product","Vectors and the Geometry of Space","Space needs three coordinates, so we set up the rectangular system, the distance formula, and the equation of a sphere. Vectors then package magnitude and direction into a single algebraic object with its own arithmetic. The dot product turns two vectors into a number that measures the angle between them, gives a clean test for orthogonality, and produces the projection of one vector onto another.\n",{"path":14544,"title":14545,"module":14541,"summary":14546},"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fthe-cross-product-lines-and-planes","The Cross Product, Lines, and Planes","The cross product multiplies two vectors into a third perpendicular to both, with length equal to the area of the parallelogram they span. That one construction supplies the direction of a line, the normal of a plane, and, through the scalar triple product, the volume of a parallelepiped. Lines carry a point and a direction vector; planes carry a point and a normal, which fixes the angle between planes and the distance from a point to a plane.\n",{"path":14548,"title":14549,"module":14541,"summary":14550},"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fcylinders-and-quadric-surfaces","Cylinders and Quadric Surfaces","A surface whose equation omits one variable is a cylinder: the graph of a plane curve swept along the missing axis. A second-degree equation in three variables is a quadric, and translation and rotation reduce every one to a short standard list. Traces — the curves cut by planes parallel to the coordinate planes — sort the six quadrics into ellipsoid, the two paraboloids, the cone, and the two hyperboloids.\n",{"path":14552,"title":14553,"module":14541,"summary":14554},"\u002Fcalculus\u002Fvectors-and-space-curves\u002Fvector-functions-and-space-curves","Vector Functions and Space Curves","A vector function assigns a vector to each value of a parameter, and as the parameter runs its tip traces a space curve. Taking limits, derivatives, and integrals component by component carries all of single-variable calculus into three dimensions. The derivative of a vector function is the tangent vector to its curve, and normalizing it gives the unit tangent that points the way along the path.\n",{"path":14556,"title":14557,"module":14541,"summary":14558},"\u002Fcalculus\u002Fvectors-and-space-curves\u002Farc-length-curvature-and-motion","Arc Length, Curvature, and Motion in Space","Integrating the speed of a vector function gives the length of its curve and a natural parameter, arc length, that depends only on the curve's shape. Curvature measures how fast the unit tangent turns, and together with the normal and binormal it builds the moving TNB frame. Reading the same vector function as a trajectory, its first two derivatives are velocity and acceleration, and acceleration splits cleanly into tangential and normal parts.\n",{"path":14560,"title":14561,"module":14562,"summary":14563},"\u002Fcalculus\u002Fpartial-derivatives\u002Ffunctions-of-several-variables","Functions of Several Variables, Limits, and Continuity","Partial Derivatives","A function of several variables assigns one number to each point of a region in the plane or in space. Domain, graph, level curve, and level surface describe it; limits and continuity extend to two variables, where a limit must agree along every path of approach, not just from the left and the right.\n",{"path":14565,"title":14562,"module":14562,"summary":14566},"\u002Fcalculus\u002Fpartial-derivatives\u002Fpartial-derivatives","A partial derivative holds every variable but one fixed and differentiates in the ordinary sense. Geometrically it is the slope of a trace curve cut from the surface by a coordinate plane. The freeze-and-differentiate rule computes the two first partials; the four second partials follow, and the two mixed ones agree under Clairaut's Theorem when they are continuous.\n",{"path":14568,"title":14569,"module":14562,"summary":14570},"\u002Fcalculus\u002Fpartial-derivatives\u002Ftangent-planes-and-the-chain-rule","Tangent Planes, Linear Approximation, and the Chain Rule","Near a point, a smooth surface looks like its tangent plane, and the plane's equation is built from the two partial derivatives. That linearization defines the total differential and the meaning of differentiability in two variables. The chain rule then propagates derivatives through composed functions, tracked by a tree diagram, and yields clean formulas for implicit differentiation.\n",{"path":14572,"title":14573,"module":14562,"summary":14574},"\u002Fcalculus\u002Fpartial-derivatives\u002Fdirectional-derivatives-and-the-gradient","Directional Derivatives and the Gradient","The partial derivatives measure slope along the two axes; the directional derivative measures slope along any chosen direction, and equals the gradient dotted with a unit vector. The gradient points in the direction of steepest increase, its length is the greatest rate, and it stands perpendicular to level curves and surfaces, which fixes the tangent plane to a level surface.\n",{"path":14576,"title":14577,"module":14562,"summary":14578},"\u002Fcalculus\u002Fpartial-derivatives\u002Foptimization-and-lagrange-multipliers","Optimization and Lagrange Multipliers","Extrema of a two-variable function sit at critical points where the gradient vanishes; the Second Derivatives Test sorts them into peaks, valleys, and saddles by the sign of a discriminant. Absolute extrema on a closed region also need the boundary. When the domain is itself a constraint curve, Lagrange multipliers set the two gradients parallel and solve the constrained problem.\n",{"path":14580,"title":14581,"module":14582,"summary":14583},"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fdouble-integrals","Double Integrals","Multiple Integrals and Vector Calculus","The double integral extends the definite integral to functions of two variables: a limit of Riemann sums that measures signed volume under a surface. Fubini's Theorem turns it into two ordinary integrations done one after the other, general regions of type I and type II fix the inner limits, polar coordinates absorb circular symmetry through the factor r, and the same machine computes mass, center of mass, and moments of a lamina.\n",{"path":14585,"title":14586,"module":14582,"summary":14587},"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Ftriple-integrals-and-coordinate-systems","Triple Integrals and Coordinate Systems","The triple integral integrates a function of three variables over a solid, as a limit of Riemann sums evaluated by three nested single integrations. Cylindrical coordinates add the factor r to handle axial symmetry, spherical coordinates add rho-squared sine-phi for radial symmetry, and the general change of variables shows both volume elements are Jacobian determinants of the coordinate map. Surface area for a graph completes the measurement toolkit.\n",{"path":14589,"title":14590,"module":14582,"summary":14591},"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fvector-fields-and-line-integrals","Vector Fields and Line Integrals","A vector field assigns a vector to every point of space; the line integral of a field along a curve accumulates its tangential component, measuring work. Conservative fields are gradients of a potential, and for them the Fundamental Theorem for Line Integrals makes the integral depend only on the endpoints. Path independence, closed-loop integrals of zero, and the component test for a potential are three faces of the same property.\n",{"path":14593,"title":14594,"module":14582,"summary":14595},"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fgreens-theorem-curl-and-divergence","Green's Theorem, Curl, and Divergence","Green's Theorem equates the line integral of a field around a positively oriented closed curve with a double integral over the enclosed region, turning a boundary computation into an area computation and vice versa. Curl measures local circulation and divergence measures local outflow; the two vector forms of Green's Theorem express the boundary integral as the integrated curl or divergence, the planar case of Stokes' and the Divergence Theorem.\n",{"path":14597,"title":14598,"module":14582,"summary":14599},"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fsurface-integrals","Parametric Surfaces and Surface Integrals","A parametric surface is the image of a two-variable vector function; its area element is the magnitude of the cross product of the two tangent vectors. The surface integral of a scalar function sums it over that area, and the flux integral of a vector field sums the field's normal component, measuring flow through the surface. Orientation by a choice of unit normal makes flux well-defined, the integral Stokes' and the Divergence Theorem operate on.\n",{"path":14601,"title":14602,"module":14582,"summary":14603},"\u002Fcalculus\u002Fmultiple-integrals-and-vector-calculus\u002Fstokes-and-the-divergence-theorem","Stokes' Theorem and the Divergence Theorem","Stokes' Theorem lifts Green's Theorem into space: the line integral of a field around the boundary of a surface equals the flux of its curl through the surface. The Divergence Theorem relates the outward flux across a closed surface to the triple integral of divergence over the solid it encloses. Together with the Fundamental Theorem of Calculus and its line-integral and Green counterparts, they are one theorem: the integral of a derivative over a region equals the integral of the field over its oriented boundary.\n",{"path":14605,"title":14606,"module":6,"summary":6},"\u002Fcalculus","Calculus",{"path":14608,"title":14609,"module":14054,"summary":14610},"\u002Fmechanics\u002Ffoundations\u002Fmeasurement-and-dimensions","Measurement and Dimensions","Every physical quantity is a number attached to a unit, and that pairing is what lets you check an equation before computing anything, since terms that add together must carry the same dimensions. We build the SI base units and the notion of dimension, then use dimensional analysis to test a proposed relation and form scaling groups — a method that fixes a formula's shape but never its numerical constants. The lesson also sets how precisely a result may be stated, through significant figures, propagated uncertainty, and order-of-magnitude checks that catch errors a raw calculator answer hides.\n",{"path":14612,"title":14613,"module":14054,"summary":14614},"\u002Fmechanics\u002Ffoundations\u002Fvector-algebra","Vector Algebra","Force, velocity, and displacement all carry a direction, so mechanics needs an arithmetic that respects it; adding magnitudes alone gives the wrong answer the moment two arrows point different ways. We set up vectors and their components in a chosen basis, then build the two products that carry most of the physics — the dot product, which extracts the part of one vector along another and yields work and power, and the cross product, which measures oriented area and yields torque and angular momentum. Rotating the axes changes the components while leaving the vector itself untouched, and the same component method resolves a force along whatever directions a constraint picks out.\n",{"path":14616,"title":14617,"module":14618,"summary":14619},"\u002Fmechanics\u002Fkinematics\u002Fone-dimensional-motion","One-Dimensional Motion","Kinematics","Motion along a line already forces the two questions the whole of kinematics repeats: how fast is the object moving now, and where will it be next? Velocity and acceleration answer the first as derivatives of position; integrating them back — the signed area under a graph — answers the second. We derive the constant-acceleration equations, mark exactly where the \"constant\" assumption is load-bearing, and see why sign, not magnitude, is what carries direction.\n",{"path":14621,"title":14622,"module":14618,"summary":14623},"\u002Fmechanics\u002Fkinematics\u002Fmotion-graphs","Motion Graphs","Draw a motion as a graph and its two most useful facts turn geometric: the slope of the position curve is the velocity, and the area under the velocity curve is the displacement. We read motion in both directions — differentiating a graph for the next rate, integrating it back to recover position — and handle the curved, piecewise, and noisy graphs that real measurements produce. Along the way we see why a velocity estimated from two positions belongs to the midpoint of their interval, not its end.\n",{"path":14625,"title":14626,"module":14618,"summary":14627},"\u002Fmechanics\u002Fkinematics\u002Fprojectile-motion","Projectile Motion","Throw an object and it seems to trace one curved path, but the motion is really two independent one-dimensional motions running at once: constant velocity across the ground and free fall in the vertical. Splitting it that way turns every projectile question — how long it stays up, how far it lands, how high it climbs, whether it clears an obstacle — into a pair of equations you already know. We derive the parabolic trajectory, work both the forward and the inverse problems, and show why the familiar $45^\\circ$ range-maximizing angle holds only when launch and landing heights match.\n",{"path":14629,"title":14630,"module":14618,"summary":14631},"\u002Fmechanics\u002Fkinematics\u002Frelative-motion","Relative Motion","A velocity is only ever measured relative to some observer, so a boat's speed through the water, over the ground, and as seen from another boat are three different vectors. Choosing the right frame — and subtracting one motion from another — collapses river crossings, crosswind headings, pursuit, and closest-approach problems into a single vector equation. We build the relative-velocity and relative-position relations for uniformly moving frames, show why acceleration is the one quantity all such observers agree on, and note where rotating frames break the simple subtraction.\n",{"path":14633,"title":14634,"module":14618,"summary":14635},"\u002Fmechanics\u002Fkinematics\u002Fcircular-motion","Circular Motion","An object going around a circle at a steady speed is still accelerating, because its velocity is forever changing direction — the fact that governs everything from a car on a curve to a satellite in orbit. We tie the angular description (angle, angular velocity, angular acceleration) to the linear one through $v=r\\omega$, split the acceleration into an inward part that turns the velocity and a tangential part that changes its speed, and extend the inward $v^2\u002Fr$ result to any curved path through its local radius of curvature. Constant angular acceleration then mirrors straight-line motion equation for equation.\n",{"path":14637,"title":14638,"module":14639,"summary":14640},"\u002Fmechanics\u002Fdynamics\u002Fnewtons-laws","Newton's Laws","Dynamics","What makes a body change its motion, and in which frames does the answer take its simplest form? Newton's three laws settle both: inertial frames are the ones where a force-free body coasts, force is whatever changes momentum, and every interaction pushes back on its source. We write the second law as $\\sum\\vec F=\\d\\vec p\u002F\\d t$, reduce it to $m\\vec a$ at constant mass, and separate what a scale actually reads — the support force — from the weight $m\\vec g$ it is so often mistaken for.\n",{"path":14642,"title":14643,"module":14639,"summary":14644},"\u002Fmechanics\u002Fdynamics\u002Ffree-body-diagrams","Free-Body Diagrams","Once several forces act on a body at once, the reliable way to predict its motion is to isolate that one body and draw every external push and pull on it — nothing more, nothing less. The free-body diagram is that discipline. We fix a system boundary, resolve $\\sum\\vec F=m\\vec a$ into components along axes chosen to fit the geometry, and solve for the unknowns a problem hands us — normal forces, tensions, friction, and the acceleration a constraint permits — seeing why internal forces drop out only when the boundary encloses both bodies that share them.\n",{"path":14646,"title":14647,"module":14639,"summary":14648},"\u002Fmechanics\u002Fdynamics\u002Ffriction-and-curved-motion","Friction and Curved Motion","Real surfaces grip before they slip, fluids push back harder the faster you move through them, and anything rounding a bend must be pulled toward the inside of the curve by something. This lesson supplies the force laws for those three cases. We bound static friction by $|f_s|\\leq\\mu_sN$ and switch to kinetic friction $\\mu_kN$ once sliding starts, model drag as a speed-dependent resistance that levels off at a terminal speed, and show that circular motion demands an inward net force $mv^2\u002Fr$ furnished by real interactions — friction, a banked normal force, tension — never by an invented outward one.\n",{"path":14650,"title":14651,"module":14639,"summary":14652},"\u002Fmechanics\u002Fdynamics\u002Fnumerical-dynamics","Numerical Dynamics","Most force laws — quadratic drag, coupled oscillators, anything nonlinear — admit no closed-form trajectory, so we advance the motion one small time step at a time and let arithmetic do what algebra cannot. This lesson turns $\\d\\vec y\u002F\\d t=f(t,\\vec y)$ into a marching rule. We derive the Euler, Euler--Cromer, midpoint, and Verlet updates, weigh their accuracy and stability, watch a drifting energy expose a bad scheme, and use step-halving and conserved quantities to separate the error of the method from the error of the model.\n",{"path":14654,"title":14655,"module":14639,"summary":14656},"\u002Fmechanics\u002Fdynamics\u002Fcenter-of-mass-systems","Center-of-Mass Systems","A firework bursts into a dozen fragments, yet one point keeps gliding along the original parabola as though nothing had happened. That point is the centre of mass, and following it collapses a many-body tangle into a single equation of motion. We define $\\vec R=\\frac1M\\sum_i m_i\\vec r_i$ and its continuous form, show that internal forces cancel so that only external ones move it, $M\\vec A_{\\rm cm}=\\sum\\vec F_{\\rm ext}$, and put the result to work on recoil, collisions viewed from the centre-of-mass frame, and rocket propulsion, where mass leaving the boundary carries momentum with it.\n",{"path":14658,"title":14659,"module":14660,"summary":14661},"\u002Fmechanics\u002Fenergy\u002Fwork-and-kinetic-energy","Work and Kinetic Energy","Energy","A constant push along a straight path is trivial to score, but real forces vary and bend along curved trajectories, and only the component along the motion transfers any energy. Work captures exactly that transfer as the line integral $W=\\int\\vec F\\cdot\\d\\vec r$, and the work-kinetic-energy theorem turns it into a statement about speed: the net work on a particle equals the change in its $\\tfrac12 mv^2$. We build work up from the dot product to the signed area under a force curve, derive the theorem from Newton's second law, and read power as its instantaneous rate $P=\\vec F\\cdot\\vec v$.\n",{"path":14663,"title":14664,"module":14660,"summary":14665},"\u002Fmechanics\u002Fenergy\u002Fpotential-energy","Potential Energy","When a force does the same work no matter which path a particle takes, that work can be stored as a function of position alone, and solving for the motion becomes bookkeeping instead of integration. We single out the forces that qualify — the conservative ones, for which $\\oint\\vec F\\cdot\\d\\vec r=0$ — define their potential energy through $\\vec F=-\\nabla U$, and use conservation of mechanical energy to read speeds, turning points, and equilibria straight off a potential curve. Friction breaks the shortcut, so we also track where mechanical energy leaks away as heat.\n",{"path":14667,"title":14668,"module":14660,"summary":14669},"\u002Fmechanics\u002Fenergy\u002Fmultiparticle-work","Multiparticle Work","A single particle has one velocity and one kinetic energy; a system of many can spin, deform, explode, and warm up while its centre of mass glides along as if nothing happened. Splitting the motion into a centre-of-mass part and an internal part separates the energy that momentum already fixes from the energy left free for relative motion, $K=\\tfrac12MV_{\\rm cm}^2+K'$. We derive the centre-of-mass work theorem, see why an explosion or a released spring can raise total kinetic energy with no external work at all, and use the reduced-mass and centre-of-mass frames to make collisions and internal transfers clean.\n",{"path":14671,"title":14672,"module":14660,"summary":14673},"\u002Fmechanics\u002Fenergy\u002Fmass-energy-and-binding","Mass-Energy and Binding","Relativity puts rest itself on the energy ledger: a mass $m$ carries energy $mc^2$ even when it sits still, so weighing a system's separated pieces and weighing the assembled whole give different answers, and the gap is binding energy. We convert freely between mass units and MeV, compute the energy that holds a nucleus together, and read the binding-energy-per-nucleon curve that explains why fusing light nuclei and splitting heavy ones both release energy. Reaction $Q$ values, thresholds, and recoil then follow from the same mass-difference accounting, once the frame and mass convention are fixed.\n",{"path":14675,"title":14676,"module":14660,"summary":14677},"\u002Fmechanics\u002Fenergy\u002Fphotons-and-quantization","Photons and Quantization","Light delivers its energy in indivisible lumps: a photon of frequency $f$ carries exactly $hf$, and this one fact explains why a dim blue lamp ejects electrons that an intense red one cannot. We fix a photon's energy and momentum from its wavelength, follow the quanta through emission, absorption, and the photoelectric threshold $K_{\\rm max}=hf-\\phi$, and watch energy and momentum conservation together produce the Compton wavelength shift when a photon scatters from an electron. The recurring discipline is unit and frame care, where a stray factor of $10^9$ or a forgotten rest energy quietly ruins an answer.\n",{"path":14679,"title":14680,"module":14681,"summary":14682},"\u002Fmechanics\u002Fmomentum\u002Fmomentum-and-collisions","Momentum and Collisions","Momentum","When two objects collide, the forces between them are too brief and too tangled to integrate directly, yet the result is fixed by one conserved quantity. Linear momentum $\\vec p=m\\vec v$ turns Newton's second law into the impulse-momentum theorem $\\vec J=\\Delta\\vec p$, and for an isolated system into a conservation law that holds through any internal collision, however dissipative. We use it to separate elastic from inelastic collisions, follow the centre of mass, and read impulse as the signed area under a force-time curve — always tracking which external impulses the chosen system and interval let us drop.\n",{"path":14684,"title":14685,"module":14681,"summary":14686},"\u002Fmechanics\u002Fmomentum\u002Fcenter-of-mass-collisions","Center-of-Mass Collisions","A two-body collision that looks asymmetric in the laboratory becomes almost trivial in the frame that rides along with the centre of mass, where the total momentum is zero and the two momenta stay equal and opposite. We build that frame, reduce the pair to a single relative coordinate carrying the reduced mass $\\mu$, and show that an elastic collision there only rotates one momentum vector while its length holds fixed. Transforming back to the laboratory then handles elastic and inelastic collisions, scattering angles, and reaction thresholds with the same construction — and shows why relative speed, not laboratory kinetic energy, measures what a collision can convert.\n",{"path":14688,"title":14689,"module":14681,"summary":14690},"\u002Fmechanics\u002Fmomentum\u002Frocket-propulsion","Rocket Propulsion","A rocket speeds up by throwing mass backward, so its own mass drops as it flies and $\\vec F=m\\vec a$ no longer applies to a fixed body. Tracking the momentum the exhaust carries across the vehicle boundary gives thrust $T=Ru_e$ and, for a force-free burn, the rocket equation $\\Delta v=u_e\\ln(m_i\u002Fm_f)$ — a logarithm that makes large velocity changes expensive in propellant and forces staging. We then add the forces a real ascent cannot ignore, gravity, drag, and steering, and show how thrust and mass-flow records are cross-checked to infer the exhaust speed.\n",{"path":14692,"title":14693,"module":14694,"summary":14695},"\u002Fmechanics\u002Frotation\u002Frotational-inertia","Rotational Inertia","Rotation","Push a wheel and a merry-go-round with the same force and they speed up at wildly different rates: the same mass resists rotation differently depending on where it sits relative to the axis. That single fact is the moment of inertia, $I=\\int r_\\perp^2\\,\\d m$, and this lesson builds it from the ground up. We tie angular motion to linear through $s=r\\theta$, $v=r\\omega$, and $a_t=r\\alpha$, derive $I$ for rods, disks, and spheres, and use the parallel- and perpendicular-axis theorems to move between axes — always naming the axis, because the same body has as many moments of inertia as it has lines to spin about.\n",{"path":14697,"title":14698,"module":14694,"summary":14699},"\u002Fmechanics\u002Frotation\u002Frotational-dynamics","Rotational Dynamics","A force applied to a wheel does nothing unless it acts off the axis: what turns a rigid body is torque, force times lever arm. This lesson makes that precise and turns it into the rotational Newton's second law, $\\sum\\tau=I\\alpha$ about a fixed axis, the exact analogue of $\\sum F=ma$. From there we get rotational work $W=\\int\\tau\\,\\d\\theta$ and power $P=\\tau\\omega$, size a motor to a load, and solve pulleys and Atwood machines where the pulley's own inertia can no longer be ignored — always insisting that every torque be measured about the same axis.\n",{"path":14701,"title":14702,"module":14694,"summary":14703},"\u002Fmechanics\u002Frotation\u002Frolling-motion","Rolling Motion","A rolling wheel is doing two things at once — translating and spinning — but the no-slip condition $v_{cm}=R\\omega$ locks them together, and that single constraint is what makes rolling tractable. We use it to split the kinetic energy into $\\tfrac12Mv_{cm}^2+\\tfrac12I\\omega^2$, find how fast a cylinder reaches the bottom of an incline, and show why the contact point is instantaneously at rest. The static friction that enforces rolling does no work; we track its direction from the tendency to slip, and mark exactly where the model breaks once the required friction exceeds $\\mu_sN$.\n",{"path":14705,"title":14706,"module":14694,"summary":14707},"\u002Fmechanics\u002Frotation\u002Fangular-momentum","Angular Momentum","A skater pulls in her arms and spins faster, with no torque acting: that is angular momentum conservation, and it lets us answer questions that would be hopeless force by force. We build $\\vec L=\\vec r\\times\\vec p$, show it obeys $\\vec\\tau_{ext}=\\d\\vec L\u002F\\d t$, and use its conservation under zero external torque to link before and after in collisions, reconfigurations, and coupled rotors without ever resolving the internal forces. The catch is bookkeeping: the origin, the system boundary, and the frame must be fixed first, and a change in total $\\vec L$ always points to an external impulse someone forgot.\n",{"path":14709,"title":14710,"module":14694,"summary":14711},"\u002Fmechanics\u002Frotation\u002Frolling-resistance","Rolling Resistance","Ideal rolling should coast forever, yet every real wheel slows down. The reason is that a deformable tire and road do not press through a single point: the contact patch spreads, the normal-force resultant shifts ahead of the axle, and that offset is a resisting moment even with no gross sliding. We package it as an equivalent force $F_{rr}=C_{rr}N$, tie the coefficient to load, surface, speed, and temperature, and use coast-down, towing, and traction tests to separate this contact loss from aerodynamic drag, bearing friction, and the adhesion limit where rolling gives way to skidding.\n",{"path":14713,"title":14714,"module":14694,"summary":14715},"\u002Fmechanics\u002Frotation\u002Fgyroscopic-precession","Gyroscopic Precession","A spinning top leans over but does not fall — it swings its axis in a slow horizontal circle instead. The paradox dissolves once torque is read as the rate of change of a vector: gravity's torque is perpendicular to the spin angular momentum, so it turns $\\vec L$ rather than toppling it. We derive the steady precession rate $\\Omega\\simeq Mgr\u002F(I_s\\omega_s)$ in the fast-top limit, state the assumptions it leans on — dominant spin, slow tilt, negligible bearing torque — and read nutation, support motion, and a decaying spin as the ways real gyroscopes depart from it.\n",{"path":14717,"title":14718,"module":14719,"summary":14720},"\u002Fmechanics\u002Fgravity-and-matter\u002Fkeplerian-orbits","Keplerian Orbits","Gravitation and Matter","Why do the planets trace ellipses rather than any other curve? Newton's inverse-square law collapses the two-body problem onto a single conic section, and the answer falls out of two conserved quantities: a central force can exert no torque, so angular momentum is fixed, and gravity is conservative, so energy is fixed. We read an orbit's size and shape straight off those invariants, recover all three of Kepler's laws, and derive escape speed, the vis-viva relation, and the timing of a pass. We also mark where the ideal ellipse breaks down — drag, oblateness, and a third body slowly move a real orbit.\n",{"path":14722,"title":14723,"module":14719,"summary":14724},"\u002Fmechanics\u002Fgravity-and-matter\u002Fgravitational-fields","Gravitational Fields","Instead of tracking the force between every pair of masses, we attach a field to the source and ask a test mass to read it off locally. That move pays off because gravity is conservative: the field is the gradient of a single scalar potential, and potentials from many sources simply add. We build the field-potential picture, use spherical symmetry and the shell theorem to get the point-mass exterior field and the zero interior field of a shell, and read tides straight out of the field's gradient. Along the way we mark exactly when the constant-$g$ and point-mass shortcuts hold and when a shape correction is needed.\n",{"path":14726,"title":14727,"module":14719,"summary":14728},"\u002Fmechanics\u002Fgravity-and-matter\u002Fstatic-equilibrium","Static Equilibrium","What does it take for a loaded structure to stay put? A body at rest needs its forces to cancel and its turning effects to cancel — $\\sum\\vec F=0$ and $\\sum\\vec\\tau=0$ about any point — and almost all of statics is the craft of turning a physical setup into those equations. We build free-body diagrams, replace supports, cables, friction, couples, and distributed loads with their idealized reactions, and locate the centre of gravity that decides whether a body tips. We also count equations against unknowns to separate a determinate problem from one that needs the material's deformation to resolve, and read every negative or inconsistent reaction as a sign that a contact or a boundary was chosen wrong.\n",{"path":14730,"title":14731,"module":14719,"summary":14732},"\u002Fmechanics\u002Fgravity-and-matter\u002Ffluid-statics","Fluid Statics","A fluid at rest cannot support a shear, so the only stress it carries is a pressure that must grow with depth to hold up the fluid above it. That single balance, $\\d p\u002F\\d z=-\\rho g$, runs the whole subject: it sets manometer readings, the force on a dam, and — integrated over a submerged boundary — Archimedes' buoyant force $F_B=\\rho g V_{\\rm disp}$. We derive these, use them to decide when a body floats and whether it floats upright, and mark where acceleration, rotation, compressibility, or capillarity forces a richer pressure model.\n",{"path":14734,"title":14735,"module":14719,"summary":14736},"\u002Fmechanics\u002Fgravity-and-matter\u002Ffluid-flow","Fluid Flow","Two accounting rules carry most of steady flow: mass cannot pile up, so the same volume crosses every section each second, and mechanical energy is conserved along a streamline when the fluid is ideal. From those we get continuity, Bernoulli's relation between pressure, speed, and height, and the results that follow — Torricelli's efflux speed, the Venturi meter, the Pitot tube. We then let go of the ideal assumptions one at a time: viscosity adds wall shear and head loss, Reynolds number decides laminar versus turbulent, and Mach number marks where a gas stops behaving as incompressible.\n",{"path":14738,"title":14739,"module":14719,"summary":14740},"\u002Fmechanics\u002Fgravity-and-matter\u002Forbital-motion","Orbital Motion","A circular orbit is nothing but free fall with enough sideways speed to keep missing the ground, and setting gravity equal to the centripetal requirement fixes that speed and the period once and for all. From the same energy bookkeeping we read off escape speed, sort orbits into bound, parabolic, and hyperbolic by the sign of their specific energy, and see why a tangential burn is the efficient way to change an orbit. We build the Hohmann transfer and its launch window, work the numbers for a geostationary orbit and an escape burn, and mark where finite thrust, perturbations, and an uncertain initial state pull a real trajectory off the ideal.\n",{"path":14742,"title":14743,"module":14719,"summary":14744},"\u002Fmechanics\u002Fgravity-and-matter\u002Fstress-and-elasticity","Stress and Elasticity","Rigid bodies are a fiction; every real material stretches, shears, or squeezes under load, and the useful question is how much. We define stress as force per area and strain as fractional deformation, then find that for small deformations the two are simply proportional — Hooke's law — with Young's, shear, and bulk moduli as the constants for stretch, twist, and volume change. From these we compute extensions, torsional twist, and stored elastic energy, and read a tensile curve for the yield, ultimate, and fracture points where linear elasticity ends. We also mark the practical limits: stress concentrations, fatigue, and the multiaxial states a single uniaxial modulus cannot capture.\n",{"path":14746,"title":14747,"module":14748,"summary":14749},"\u002Fmechanics\u002Foscillations-waves\u002Fdamped-oscillators","Damped Oscillators","Oscillations and Waves","Every real oscillator eventually stops: friction, drag, and internal loss drain its energy, so free motion is a decay rather than a permanent swing. Adding a velocity-proportional resistance to the spring-mass equation produces one dimensionless number, $b\u002F(2\\sqrt{mk})$, that decides whether the mass rings down through many cycles, returns once without overshoot, or crawls back slowly. We solve the three regimes, tie the observed decay to the power balance $b\\dot x^2$, and turn a measured ring-down into the decay rate and quality factor of the apparatus — reading damping off the data instead of assuming it.\n",{"path":14751,"title":14752,"module":14748,"summary":14753},"\u002Fmechanics\u002Foscillations-waves\u002Ftravelling-waves","Travelling Waves","A wave carries a shape, not the material: each element of a rope or air column oscillates in place while the disturbance travels through it. Writing that shape as $f(x\\mp vt)$ turns \"the pattern moves\" into a statement about the cosine's argument, and a local force balance on one string segment fixes the speed at $v=\\sqrt{T\u002F\\mu}$ — restoring stiffness over inertia, with amplitude nowhere in it. We build the sinusoidal wave and its phase, derive the wave equation from Newton's second law, and follow the energy a travelling wave transports, then check speed and power against those predictions.\n",{"path":14755,"title":14756,"module":14748,"summary":14757},"\u002Fmechanics\u002Foscillations-waves\u002Fwave-superposition","Wave Superposition","When two waves cross the same point, what does a probe read? In a linear medium the answer is arithmetic: the displacements add, $y=y_1+y_2$, and the pulses pass through each other unchanged. That one rule produces interference — reinforcement where the signs agree, cancellation where they oppose — and it guards against a common mistake, since displacement can vanish at an instant while the energy sits in transverse motion instead. We work out the signed sum, the phase bookkeeping for equal-frequency components, and why a null in the record is not a null in the wave.\n",{"path":14759,"title":14760,"module":14748,"summary":14761},"\u002Fmechanics\u002Foscillations-waves\u002Fstanding-waves","Standing Waves","Clamp a string at both ends and only certain frequencies survive: the ends must be nodes, and that single geometric demand quantizes the wave into a discrete set of modes $f_n=nv\u002F(2L)$. The travelling wave becomes a fixed pattern of nodes and antinodes — standing, not moving — because equal waves running in opposite directions superpose. We build the standing wave from its counter-propagating pieces, read the harmonic sequence off the boundary conditions (half-wavelengths for a fixed-fixed string, odd quarter-wavelengths for a closed pipe), and test the ideal model against node scans and resonance peaks.\n",{"path":14763,"title":14764,"module":14748,"summary":14765},"\u002Fmechanics\u002Foscillations-waves\u002Fsound-waves","Sound Waves","Sound is a pressure wave so small that a loud tone displaces air molecules by less than the width of an atom, yet a microphone reads it easily — because pressure, not displacement, is what the ear and the instrument sense. The acoustic impedance $Z=\\rho c$ ties pressure, density, and particle velocity together, fixes the intensity a wave carries, and sets the reference for the decibel, a logarithm that tames a $10^{12}$ range in power. We derive the sound speed from the gas's stiffness, convert between pressure and intensity levels, and treat the measurement itself — calibration, geometry, background, averaging — as part of the physics.\n",{"path":14767,"title":14768,"module":14748,"summary":14769},"\u002Fmechanics\u002Foscillations-waves\u002Fdoppler-effect","Doppler Effect","A passing siren drops in pitch not because the source changes but because motion repacks the wavefronts: an approaching source crowds its crests, a receding one stretches them, and a moving listener samples them at a different rate. For mechanical waves every velocity is measured against the medium, and one signed ratio $f_r=f_s(v-u_r)\u002F(v-u_s)$ captures both effects at once. We separate source motion, which sets crest spacing, from receiver motion, which sets arrival rate, invert the shift to recover radial velocity, and mark where the model breaks — supersonic sources, moving air, and reflected paths that carry two shifts, not one.\n",{"path":14771,"title":14772,"module":14748,"summary":14773},"\u002Fmechanics\u002Foscillations-waves\u002Fwave-packets","Wave Packets","No real signal is a single frequency: a disturbance that starts and stops is built from a band of wave numbers, and the width of that band is what makes it local. We ask how such a packet moves — carrier crests at the phase velocity $v_\\mathrm p=\\omega\u002Fk$, the envelope at the group velocity $v_\\mathrm g=\\d\\omega\u002F\\d k$ — and why the two differ once a medium is dispersive. Curvature $\\d^2\\omega\u002F\\d k^2$ spreads and chirps the packet as it travels, and the Fourier reciprocity that ties bandwidth to duration explains why a finite record, aliasing, or a coarse probe can imitate that spreading unless the sampling limits are respected.\n",{"path":14775,"title":14776,"module":14748,"summary":14777},"\u002Fmechanics\u002Foscillations-waves\u002Fbeats-and-coupling","Beats and Coupling","Add two tones a few hertz apart and the sum swells and fades at their difference frequency — a beat — though neither source is changing. We work out that envelope, then ask the mechanical version of the same question: join two oscillators and a single resonance splits into normal modes, with energy sloshing between the coordinates at their frequency difference. The lesson identifies when a slow amplitude envelope signals genuine coupling rather than two independent sources, drift, or deliberate modulation, reading it from envelope timing, spectral sidebands, and the mode shapes.\n",{"path":14779,"title":14780,"module":14748,"summary":14781},"\u002Fmechanics\u002Foscillations-waves\u002Fsimple-harmonic-motion","Simple Harmonic Motion","Any system pushed back toward equilibrium by a force proportional to its displacement obeys one equation, $\\ddot x+\\omega_0^2x=0$, and so moves sinusoidally at $\\omega_0=\\sqrt{k\u002Fm}$ whatever the amplitude. We derive that motion, follow its energy $E=mv^2\u002F2+kx^2\u002F2$ trading between kinetic and potential form at constant total, and read the elliptical phase-space orbit Hooke's law implies. Period, amplitude, velocity, and acceleration then supply redundant checks: an amplitude-dependent period or a curved force residual is the signature that the linear model has failed, and mass-loading and offset tests separate a calibration error from a real frequency shift.\n",{"path":14783,"title":14784,"module":14748,"summary":14785},"\u002Fmechanics\u002Foscillations-waves\u002Fpendulum-motion","Pendulum Motion","A pendulum keeps time only because, for small swings, gravity supplies a restoring torque proportional to the angle — and $T=2\\pi\\sqrt{L\u002Fg}$ then follows without the mass appearing at all. We derive that result, mark exactly which assumptions carry it (small angle, negligible pivot loss, a rigid support), then relax them: finite amplitude lengthens the period through an elliptic integral, and an extended body replaces $L$ with the ratio of its moment of inertia to its center-of-mass distance. How the period drifts with amplitude or pivot position is what diagnoses the geometric, damping, and distributed-mass corrections.\n",{"path":14787,"title":14788,"module":14748,"summary":14789},"\u002Fmechanics\u002Foscillations-waves\u002Fdriven-oscillators","Driven Oscillators","Drive a damped oscillator at a frequency you control and it eventually forgets its own: $m\\ddot x+b\\dot x+kx=F_0\\cos\\omega t$ settles into a steady response whose amplitude and phase depend sharply on how close the drive sits to resonance. We solve for that response, show how damping alone fixes the resonance width, the peak power, and the settling time, and treat base excitation as the same problem with a different input. The steady-state formulas hold only for constant $m$, $b$, and $k$; level-dependent peaks or hysteresis between up- and down-sweeps are how nonlinearity or an extra mode announces itself.\n",{"path":14791,"title":14792,"module":14748,"summary":14793},"\u002Fmechanics\u002Foscillations-waves\u002Fwave-boundaries","Wave Boundaries","A pulse traveling along a string does something abrupt where the string's properties change: part reflects, part transmits, and which is which is set by the impedance mismatch alone. We impose continuity of displacement and transverse force at the join to get the reflection and transmission coefficients in terms of $Z=\\sqrt{T\\mu}$, fix their signs and the polarity flip, and balance the energy. The clean result assumes linear, nondispersive segments meeting at a localized join; pulse polarity, return timing, and energy ratios are the measurements that expose a real connector's mass, loss, or distributed transition.\n",{"path":14795,"title":14796,"module":14797,"summary":14798},"\u002Fmechanics\u002Fthermodynamics\u002Fkinetic-theory-of-ideal-gases","Kinetic Theory of Ideal Gases","Thermodynamics","A gas has no springs and no gears, yet it pushes on its container with a definite pressure and stores energy in a lawful way. Kinetic theory explains both from the motion of the molecules alone: pressure is the accumulated recoil of countless elastic impacts, and temperature is the average translational kinetic energy each molecule carries. We derive $pV=\\tfrac13Nm\\overline{v^2}$ from momentum transfer, read off $\\overline{K}_{\\rm tr}=\\tfrac32kT$, and use the Maxwell–Boltzmann distribution to separate the most probable, mean, and rms speeds — each the right average for a different question — while marking where the dilute, classical assumptions stop holding.\n",{"path":14800,"title":14801,"module":14797,"summary":14802},"\u002Fmechanics\u002Fthermodynamics\u002Ffirst-law-of-thermodynamics","First Law of Thermodynamics","Heat a gas and it may warm, expand, or both; compress it and the same energy can reappear as a temperature rise. The first law settles the bookkeeping: internal energy is a state property whose change equals the heat added plus the work done on the system, $\\Delta E_{\\rm int}=Q_{\\rm in}+W_{\\rm on}$. We fix a system boundary and one sign convention, compute boundary work as $\\int p\\,\\d V$ along a path, and use calorimetry to measure heat and heat capacities. The recurring point is that heat and work are path-dependent transfers while their sum is not, so an energy ledger closes only once every boundary crossing is named.\n",{"path":14804,"title":14805,"module":14797,"summary":14806},"\u002Fmechanics\u002Fthermodynamics\u002Fentropy-and-the-second-law","Entropy and the Second Law","The first law lets energy flow either way; it never says which way heat actually goes. The second law supplies the missing arrow. Entropy, defined through the reversible transfer $\\d S=\\delta Q_{\\rm rev}\u002FT$, can only increase in an isolated system, and that single inequality fixes the direction of heat flow and caps every engine, refrigerator, and heat pump at its Carnot value. We build entropy ledgers for reservoirs and working substances, separate the entropy carried by heat from the entropy generated by irreversibility, and read the sign of the total as a hard check on any proposed thermal machine.\n",{"path":14808,"title":14809,"module":14797,"summary":14810},"\u002Fmechanics\u002Fthermodynamics\u002Fthermal-processes","Thermal Processes","Heat rarely sits still: it stretches solids, pushes real gases off their ideal isotherms, and leaks across walls by conduction, convection, and radiation. Each behavior becomes a number a designer can use. Thermal expansion sets the gaps in a bridge and the stress in a clamped rod; the van der Waals equation and a phase diagram fix when $pV=nRT$ or a latent-heat term applies; Fourier's law, Newton cooling, and Stefan–Boltzmann radiation give the rate of heat flow. We assemble these into thermal-resistance networks and transient time constants, then mark where contact resistance, phase change, or a hidden thermal bridge breaks the simple model.\n",{"path":14812,"title":14813,"module":14797,"summary":14814},"\u002Fmechanics\u002Fthermodynamics\u002Fphase-changes","Phase Changes","Add heat to ice and its temperature climbs — until it reaches $0\\ ^\\circ\\mathrm C$, where the thermometer stalls while the ice melts. That plateau is the whole subject: at a phase boundary the energy rearranges molecules, $Q=mL$, instead of raising temperature, which resumes only once one phase is gone. We stage a heating path into sensible-heat legs ($Q=mc\\Delta T$) and latent plateaus, use the Clausius–Clapeyron relation to track how a boiling point moves with pressure, and solve calorimetry by testing each coexistence endpoint — so a melt fraction that lands outside $[0,1]$ flags a wrong final-state guess rather than a real state.\n",{"path":14816,"title":14817,"module":14797,"summary":14818},"\u002Fmechanics\u002Fthermodynamics\u002Fthermal-machines","Thermal Machines","An engine, a refrigerator, and a heat pump are one machine read three ways: each shuttles heat between a hot and a cold reservoir while trading work at the boundary, and only the flow you call useful separates them. A heat engine turns part of $Q_h$ into work, $W=Q_h-Q_c$; a refrigerator spends work to pull $Q_c$ from the cold side; a heat pump counts the warm-side delivery instead. We measure each with its own ratio — efficiency or coefficient of performance — bound them all by the Carnot limit that reservoir temperatures alone set, and track how finite temperature differences, throttling, and friction generate entropy and pull real machines below that bound.\n",{"path":14820,"title":14821,"module":6,"summary":6},"\u002Fmechanics","Mechanics & Dynamics",{"path":14823,"title":14824,"module":14825,"summary":14826},"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Fcharge-and-conductors","Charge and Conductors","Electric Fields","Rub two objects together and one pulls electrons from the other; nothing is created, only moved. We define what electric charge is — conserved, additive, and quantized in units of $e$ — and why a conductor's mobile carriers rearrange until its interior field vanishes and its surface sits at one potential. We follow charge through contact, induction, and grounding, treat the field-free cavity that turns a conductor into a shield, and mark where finite conductivity and leakage set the limits of the electrostatic picture.\n",{"path":14828,"title":14829,"module":14825,"summary":14830},"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Fcoulombs-law","Coulomb's Law","Two charges at rest push or pull along the line joining them, and the whole of electrostatics is assembled by adding up such pairs. We measure that force — its inverse-square falloff, its linear dependence on each charge, the sign that says attract or repel — and write it as a vector so direction survives superposition. We work the magnitude and component forms on real numbers, check them against limiting cases and dimensions, and fix the point-charge approximation to source sizes small against every separation.\n",{"path":14832,"title":14833,"module":14825,"summary":14834},"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Felectric-field-and-force","Electric Field and Force","Rather than ask how one charge reaches across empty space to another, we credit the source with a field that fills the space and let a second charge respond to whatever field sits at its own location. Electric field is force per unit positive test charge, $\\vec E=kq\\hat r\u002Fr^2$ for a point source, and source fields add before any receiving charge is placed. We compute those fields and the force $\\vec F=q\\vec E$ they exert, then follow a charge along its parabolic path through a uniform field and into nonuniform fields where the dynamics turn position-dependent.\n",{"path":14836,"title":14837,"module":14825,"summary":14838},"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Felectric-field-maps","Electric Field Maps","A field is a vector at every point of space, and the quickest way to grasp one is to draw it. We build the two standard pictures — continuous field lines tangent to $\\vec E$, and scaled vector arrows — and read direction, magnitude, and the location of nulls straight off them. We fix what a line drawing can and cannot say: density encodes magnitude only under a stated seeding rule, and integral curves never cross at a regular point. From there we work the topology near sources, sinks, and conductor surfaces, and state the step-size and interpolation checks a numerical map must pass.\n",{"path":14840,"title":14841,"module":14825,"summary":14842},"\u002Felectricity-and-magnetism\u002Felectric-fields\u002Felectric-dipoles","Electric Dipoles","Most neutral matter carries no net charge yet still responds to an electric field, because its positive and negative charge sit slightly apart. That separation is a dipole, moment $\\vec p=q\\vec d$ pointing from the negative to the positive charge, and it is the leading term in how any neutral distribution looks from far away. We derive the torque $\\vec p\\times\\vec E$ and energy $-\\vec p\\cdot\\vec E$ a uniform field imposes, the net force a field gradient adds, and the axial and equatorial $1\u002Fr^3$ fields the pair produces, then measure how far out the point-dipole approximation still holds.\n",{"path":14844,"title":14845,"module":14846,"summary":14847},"\u002Felectricity-and-magnetism\u002Fcontinuous-charge-distributions\u002Fcontinuous-charge-fields","Continuous Charge Fields","Continuous Charge Distributions","A charged rod, ring, or disk is not a point, yet its field is still nothing but Coulomb's law added up over the charge it carries. We replace the discrete sum by an integral, with $\\d q=\\lambda\\d\\ell$, $\\sigma\\d A$, or $\\rho\\d V$, so the real work becomes geometry: writing the vector from each source element to the field point, and letting symmetry cancel the components that must cancel before any integral is attempted. We carry the line, ring, and disk fields through in full, then check each result against its near field, its far field, and its dimensions.\n",{"path":14849,"title":14850,"module":14846,"summary":14851},"\u002Felectricity-and-magnetism\u002Fcontinuous-charge-distributions\u002Fgauss-law-and-conductors","Gauss's Law and Conductors","Adding up Coulomb's law over a whole distribution is laborious; Gauss's law trades that sum for a single statement, that the flux of $\\vec E$ out of any closed surface counts the charge inside, $\\oint\\vec E\\cdot\\d\\vec A=Q_{\\rm enc}\u002F\\varepsilon_0$. The law is always true, but it hands over the field only when the source is symmetric enough to pull $E$ outside the integral. We apply it to spheres, lines, and sheets, then turn it on conductors, where the zero interior field drives every excess charge to the surface and fixes the normal-field jump $\\sigma\u002F\\varepsilon_0$, the charge induced on a cavity wall, and electrostatic shielding.\n",{"path":14853,"title":14854,"module":14855,"summary":14856},"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Fpoint-charge-potential","Point-Charge Potential","Electric Potential","The electrostatic force is conservative, so the work it does between two points\ndepends only on the endpoints. That lets us trade the vector field for a single\nscalar attached to each point, the electric potential, the potential energy a unit\ncharge would have there. We build potential from the work integral, fix the usual\nreference at infinity, and add point sources as scalars, $V=k\\sum_i q_i\u002Fr_i$,\navoiding the vector bookkeeping the field demands. Signed charges, the reference\nchoice, equipotential motion, and far-field expansions each give an independent\ncheck on a result.\n",{"path":14858,"title":14859,"module":14855,"summary":14860},"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Fpotential-gradients-and-equipotentials","Potential Gradients and Equipotentials","Given the potential everywhere, how do we recover the field? The field is the\nnegative gradient, $\\vec E=-\\nabla V$: it points down the steepest local drop in\npotential, and its magnitude is set by how fast $V$ changes, not by the shape of a\ncontour. We read off components with directional derivatives, reconstruct fields\nfrom measured potential grids using centered differences, and use closed-loop\nintegrals and grid refinement to test whether a reconstructed field is physically\nconsistent.\n",{"path":14862,"title":14863,"module":14855,"summary":14864},"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Felectrostatic-energy-and-pressure","Electrostatic Energy and Pressure","Assembling a charge configuration takes work, and that work is stored, but where\nis it kept and how much is there? We total it two ways: as a sum over the charges,\n$U=\\tfrac12\\sum_i q_iV_i$, and as an integral over the field itself,\n$u_E=\\tfrac12\\varepsilon_0E^2$, energy the field carries in every region it fills.\nDifferentiating the stored energy at fixed charge or at fixed voltage recovers the\nmechanical force on a conductor, and at a charged surface the same field scale\nappears as an outward electrostatic pressure. We work the parallel-plate case in\nfull and mark where curvature and fringing make the pressure nonuniform.\n",{"path":14866,"title":14867,"module":14855,"summary":14868},"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Flaplace-boundary-problems","Laplace Boundary Problems","Often the charges are not given, only the conductors and the voltages held on\nthem, and the potential in the empty space between has to be found. There $V$ obeys\nLaplace's equation $\\nabla^2V=0$, and the boundary data alone determine a unique solution.\nWe solve it two ways: separation of variables into boundary-matched modes, whose\nhigher spatial frequencies die away with depth into the domain, and finite-difference\nrelaxation for boundaries no analytic mode fits. Residual norms, boundary error, and\nflux balance tell us when the computed potential and its field can be trusted.\n",{"path":14870,"title":14871,"module":14855,"summary":14872},"\u002Felectricity-and-magnetism\u002Felectric-potential\u002Fcontinuous-charge-potentials","Continuous Charge Potentials","When charge is spread over a line, a surface, or a volume, the sum over point\nsources becomes an integral, $V(\\vec r)=k\\int \\d q\u002F|\\vec r-\\vec r'|$. Because\npotential is a scalar, this integral sidesteps the component algebra the field\nwould force, until the field is actually wanted through $\\vec E=-\\nabla V$. We set\nup the right density element for each geometry, choose a workable reference, handle\nthe integrable singularities that arise when the observation point sits on the\ncharge, and check every result against symmetry, dimensions, and the far-field\nmultipole limit.\n",{"path":14874,"title":14875,"module":14876,"summary":14877},"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fcapacitance-fundamentals","Capacitance Fundamentals","Capacitance","How much charge must you separate onto two conductors to hold a given voltage between\nthem? That ratio, $C=Q\u002F\\Delta V$, is fixed by the conductor geometry and the medium,\nnot by how much charge is presently stored. We compute it from the field for the\nparallel-plate, isolated-sphere, concentric-sphere, and coaxial geometries, trace how\nsurface charge and boundary conditions set each result, and see where fringing,\nguarding, and stray coupling separate the ideal formula from what a bridge measures.\n",{"path":14879,"title":14880,"module":14876,"summary":14881},"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fcapacitor-networks","Capacitor Networks","Wire several capacitors together and the source sees one equivalent capacitance — but\nwhich? The answer comes not from how the symbols are drawn but from which conductors\nshare a node: parallel branches hold a common voltage and add, $C_{\\rm eq}=\\sum_iC_i$,\nwhile series branches share a common charge and add reciprocally. We derive both rules\nfrom charge conservation on the floating internal node, then extend the node-charge\nmethod to unequal, precharged, and stray-coupled branches and carry a worked reduction\nthrough to the charge and voltage on every element.\n",{"path":14883,"title":14884,"module":14876,"summary":14885},"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fcapacitor-energy-and-force","Capacitor Energy and Force","Charging a capacitor takes work, because every increment of charge is pushed through\nthe voltage the earlier charge already established. We total that work three\nequivalent ways, $U=Q^2\u002F(2C)=Q\\Delta V\u002F2=C(\\Delta V)^2\u002F2$, locate it in the field as\na density $u=\\tfrac12\\epsilon_0E^2$, then let the plates move. Differentiating the\nstored energy at fixed charge, or the coenergy at fixed voltage, gives the mechanical\nforce; the two boundaries differ only by the work the source supplies. We work the\nparallel-plate attraction and its electrostatic pressure in full, and follow the same\ngradient into pull-in, tilt, comb drives, and traceable force calibration.\n",{"path":14887,"title":14888,"module":14876,"summary":14889},"\u002Felectricity-and-magnetism\u002Fcapacitance\u002Fdielectric-polarization-and-breakdown","Dielectric Polarization and Breakdown","Slide a dielectric between the plates and the capacitance rises — but why, and how\nhard can you drive it before the insulator fails? Bound charge answers the first:\npolarization $\\vec P$ sets up surface and volume charge that partly cancels the\napplied field, so $\\vec D=\\varepsilon_0\\vec E+\\vec P$ separates what the circuit\ncontrols from what the material contributes. We follow the field across layered\ndielectrics and interfaces, tie permittivity and loss to their frequency dependence,\nand treat dielectric strength as a measured, geometry-dependent limit rather than one\nmaterial number.\n",{"path":14891,"title":14892,"module":14893,"summary":14894},"\u002Felectricity-and-magnetism\u002Fdirect-current-circuits\u002Fcurrent-and-resistance","Current and Resistance","Direct-Current Circuits","What does it mean, physically, for charge to flow, and what sets how hard a wire resists that flow? Current counts charge crossing a surface, $I=\\int\\vec J\\cdot\\d\\vec A$, and traces back to a slow drift of many carriers, $\\vec J=nq\\vec v_d$. We establish when the linear law $V=IR$ actually holds, how resistivity and geometry combine into bulk resistance, why real sources sag under load through their internal resistance, and how the three power forms $P=IV=I^2R=V^2\u002FR$ tie electrical work to heating and component ratings.\n",{"path":14896,"title":14897,"module":14893,"summary":14898},"\u002Felectricity-and-magnetism\u002Fdirect-current-circuits\u002Fkirchhoff-network-analysis","Kirchhoff Network Analysis","Once a circuit has more than one loop, no amount of series-parallel folding will reduce it — you need the two conservation laws written as equations. Kirchhoff's junction law is charge conservation at a node; his loop law is energy conservation around a closed path. We turn a labelled network into a linear system in node voltages or mesh currents, fix the sign conventions so a negative answer just means a reversed arrow, and use power balance as an independent check that the algebra describes the circuit that was actually built.\n",{"path":14900,"title":14901,"module":14893,"summary":14902},"\u002Felectricity-and-magnetism\u002Fdirect-current-circuits\u002Frc-transients","RC Transients","How does a circuit get from one steady state to the next when a capacitor refuses to change its voltage all at once? Because a jump would demand infinite current, an RC circuit slides between states exponentially, with a single time constant $\\tau=RC$ that sets the whole schedule: charging fills as $1-e^{-t\u002F\\tau}$, discharge empties as $e^{-t\u002F\\tau}$. We solve the first-order loop equation, read the response off three numbers — the switch-instant voltage, the final dc voltage, and the Thevenin resistance the capacitor sees — and mark where source and probe resistance shift $\\tau$ or where a second storage element hides a mode a one-$\\tau$ fit misses.\n",{"path":14904,"title":14905,"module":14906,"summary":14907},"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmagnetic-trajectories","Magnetic Trajectories","Magnetic Field","A charged particle in a magnetic field never speeds up or slows down, yet its path curves relentlessly. We work out why: the magnetic force is always perpendicular to velocity, so it does no work and bends the transverse motion into a circle of radius $r=mv_\\perp\u002F(|q|B)$ while leaving the parallel motion untouched, producing a helix. We derive the cyclotron frequency, show why it is independent of speed until relativity intervenes, and turn the geometry around: a measured curvature reads back a particle's momentum, which is how tracking detectors weigh what they cannot see.\n",{"path":14909,"title":14910,"module":14906,"summary":14911},"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fhall-effect","Hall Effect","Current tells you charge is moving, but not whether the movers are positive or negative, nor how many there are. A magnetic field settles both questions. Push current through a strip in a transverse field and the carriers pile up on one edge until a transverse electric field just balances the magnetic deflection; the sign of the resulting Hall voltage names the carrier's charge and its size counts the carriers per volume. We derive the balance $q\\vec E+q\\vec v_d\\times\\vec B=0$, read off $V_H=IB\u002F(nqt)$, and see why field-and-current reversal is what separates the real Hall signal from the offsets that mimic it.\n",{"path":14913,"title":14914,"module":14906,"summary":14915},"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmagnetic-force-on-conductors","Magnetic Force on Conductors","A magnet pushes on a current-carrying wire even though the wire is electrically neutral. The reason is that each moving carrier feels the Lorentz force, and those microscopic pushes add up to a force the wire's supports must hold. We sum them into $\\d\\vec F=I\\,\\d\\vec\\ell\\times\\vec B$, collapse it to $\\vec F=I\\vec L\\times\\vec B$ for a straight segment in a uniform field, and see exactly when that shortcut fails and the full path integral is needed. The same law runs backward as a measurement: a force-versus-current slope weighs a magnetic field against a known length.\n",{"path":14917,"title":14918,"module":14906,"summary":14919},"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmagnetic-dipoles","Magnetic Dipoles","A compass needle turns to point north; a current loop in a field does the same thing, and for the same reason. Both are magnetic dipoles, and a uniform field cannot push a dipole anywhere, only twist it. We package a loop's response into one vector, the magnetic moment $\\vec\\mu=IA\\hat n$, from which torque $\\vec\\tau=\\vec\\mu\\times\\vec B$ and orientation energy $U=-\\vec\\mu\\cdot\\vec B$ both follow. Stable alignment sits at the energy minimum, a field gradient is what it takes to produce a net force $\\vec F=\\nabla(\\vec\\mu\\cdot\\vec B)$, and the same moment reappears whenever anything from an electron to a planet acts magnetic.\n",{"path":14921,"title":14922,"module":14906,"summary":14923},"\u002Felectricity-and-magnetism\u002Fmagnetic-field\u002Fmass-spectrometry","Mass Spectrometry","To weigh a single atom you cannot use a scale, so you use a magnetic field instead. A charged ion of unknown mass bends in a field by an amount that depends on its momentum and charge, so if every ion enters with the same velocity, its landing position reads off its mass-to-charge ratio directly. We build the instrument in two stages: crossed electric and magnetic fields that pass only ions with $v=E\u002FB$, and a magnetic sector that bends the survivors along $r=mv\u002F(|q|B)$. Then we ask what blurs a spectral line and how reference ions turn a position into a calibrated mass.\n",{"path":14925,"title":14926,"module":14927,"summary":14928},"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fmoving-charge-fields","Moving-Charge Fields","Magnetic Sources","Every magnetic field comes from charge in motion, and the simplest source is a single point charge drifting past. We work out the field it produces — normal to both the velocity and the line of sight, falling off as the inverse square — and read off why it vanishes straight ahead of the charge and peaks broadside. Summing many such charges is the bridge to steady currents, valid while speeds stay far below $c$ and the motion changes little during the time its field takes to propagate outward.\n",{"path":14930,"title":14931,"module":14927,"summary":14932},"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fbiot-savart-law","Biot–Savart Law","A steady current is a continuous stream of current elements, and the Biot–Savart law hands each one a magnetic contribution — a right-hand cross product that falls off as the inverse square of distance. Summing the contributions along a conductor is a vector line integral, which we carry out for the straight wire to get the endpoint-angle formula. The infinite-wire field $B=\\mu_0 I\u002F2\\pi s$ falls out as the limit where both ends recede, and we mark how fast a finite wire departs from it and when a thin-filament model is safe.\n",{"path":14934,"title":14935,"module":14927,"summary":14936},"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fcircular-current-loops","Circular Current Loops","A ring of current is the simplest source with a well-defined magnetic axis, and it is the building block of every coil and electromagnet. Symmetry kills the transverse Biot–Savart contributions along that axis and leaves a single clean integral; we evaluate it to get $B_z=\\mu_0 I R^2\u002F[2(R^2+z^2)^{3\u002F2}]$, read off the centre field $\\mu_0 I\u002F2R$, and watch it fall into the $1\u002Fz^3$ tail of a magnetic dipole far away. Stacking turns just adds their axial contributions, which is what makes a solenoid out of a pile of loops.\n",{"path":14938,"title":14939,"module":14927,"summary":14940},"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Famperes-law","Ampère’s Law","When a current arrangement is symmetric enough, the Biot–Savart integral is overkill: Ampère's law, $\\oint_C\\vec B\\cdot\\d\\vec\\ell=\\mu_0 I_{\\rm enc}$, gets the field from a single line of reasoning about how much current a loop encloses. We see why the law holds for any steady current, then use cylindrical, planar, and toroidal symmetry to turn the circulation into simple algebra — the field inside and outside a wire, an infinite sheet, a solenoid, and a toroid. We also mark the catch: without symmetry the law still holds but no longer hands you the field pointwise.\n",{"path":14942,"title":14943,"module":14927,"summary":14944},"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fgauss-law-for-magnetism","Gauss’s Law for Magnetism","Electric field lines start and end on charges; magnetic field lines do neither, because no one has ever found an isolated magnetic pole. That single experimental fact is Gauss's law for magnetism: the flux of $\\vec B$ through any closed surface is zero, $\\oint\\vec B\\cdot\\d\\vec A=0$, or in differential form $\\nabla\\cdot\\vec B=0$. We work through what it says — every field line that enters a closed surface must leave it, so field lines close on themselves — and, just as important, what it does not say, since flux through an open surface is generally nonzero.\n",{"path":14946,"title":14947,"module":14927,"summary":14948},"\u002Felectricity-and-magnetism\u002Fmagnetic-sources\u002Fmagnetic-materials","Magnetic Materials","Put matter in a magnetic field and its atoms respond, each acting as a tiny current loop; the aligned moments per unit volume are the magnetization $\\vec M$, whose bound currents add to the field. Separating what we control (the free current) from what the material supplies leads to $\\vec H$ and the relation $\\vec B=\\mu_0(\\vec H+\\vec M)$. We sort materials into diamagnets, paramagnets, and ferromagnets by how $\\vec M$ answers, follow a ferromagnet around its hysteresis loop, and see why the loop's area is the energy dissipated per cycle and why a sample's shape changes the field it actually feels.\n",{"path":14950,"title":14951,"module":14952,"summary":14953},"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fmagnetic-flux","Magnetic Flux","Electromagnetic Induction","A magnetic field threading a loop collapses to one signed number, the flux, and every induced voltage in this module turns out to be a rate of change of that number — so defining the flux and its sign comes first. We define it as the surface integral of $\\vec B$ over an oriented surface, reduce it to $BA\\cos\\theta$ for a uniform field on a flat loop, and carry the flux linkage $N\\Phi_B$ of a coil. The chosen normal fixes the sign; reversing it flips the sign without touching the field. Nonuniform fields and curved surfaces force the integral, so we also build the numerical estimate and the checks that separate a reliable value from a nominal field-times-area product.\n",{"path":14955,"title":14956,"module":14952,"summary":14957},"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Ffaradays-law","Faraday's Law","Move a magnet toward a coil, or ramp the current in a nearby circuit, and a voltage appears with no battery in sight. Faraday's law names the cause: the emf around a loop equals minus the rate of change of the magnetic flux through it, so any change of field, area, orientation, or position that alters the flux drives an emf. We separate the emf, which lives around the boundary whether or not current can flow, from the current that follows only when the path is closed; fix the single sign convention that ties flux to loop orientation; and read the emf off rotating coils and off flux sampled at discrete times.\n",{"path":14959,"title":14960,"module":14952,"summary":14961},"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Flenzs-law","Lenz's Law","The minus sign in Faraday's law is not decoration: it decides which way the induced current flows, and it always chooses the direction that fights the change that produced it. Lenz's law reads that sign off energy conservation — a current that aided the change would be free energy — and turns it into a repeatable procedure. We fix a surface normal and a positive loop direction so the sign is calculable, then work through approaching magnets, expanding loops, coupled coils, and rotating generators, using mechanical work and Joule heating as an independent check on every direction we draw.\n",{"path":14963,"title":14964,"module":14952,"summary":14965},"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fmotional-emf","Motional EMF","Push a wire through a magnetic field and its free charges feel a sideways magnetic force that piles them up at the ends — a battery made of motion. Motional emf is that effect: the work per unit charge a moving conductor supplies is the line integral of $\\vec v\\times\\vec B$ along it, which for a rod moving perpendicular to both its length and the field collapses to $B\\ell v$. We chase where the energy comes from — the hand or motor fighting the magnetic drag, never the magnetic force itself — solve the sliding-rail circuit from both flux and carrier forces, and carry the idea into rotating rods, homopolar disks, generators, and the back emf of a motor.\n",{"path":14967,"title":14968,"module":14952,"summary":14969},"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Feddy-currents","Eddy Currents","A wire carries current along one path; a solid block of metal offers a continuum of them, and any changing flux threading that block sets charge circulating in closed loops it chooses for itself. We ask what those eddy currents do — where they heat, where they drag, and how Lenz's law fixes their direction — and why the same circulation is a feature in an induction furnace and a loss to be suppressed in a transformer core. From a representative-loop estimate we get the scaling (heating grows with the square of frequency and flux rate) and the two design levers, lamination and resistivity, that break the paths a solid conductor would otherwise hand the current.\n",{"path":14971,"title":14972,"module":14952,"summary":14973},"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fself-inductance","Self-Inductance","A coil resists changes to its own current. Drive current through it and the flux it produces threads its own turns; change that current and Faraday's law turns the coil against the source with a back emf $\\mathcal E_L=-L\\,\\d I\u002F\\d t$. We define self-inductance as the flux linkage per ampere fixed by winding and core geometry, derive the long-solenoid value $L=\\mu_0 N^2A\u002F\\ell$, and follow the consequence that dominates circuits: because a finite voltage can only sustain a finite $\\d I\u002F\\d t$, an inductor's current cannot jump — which is why opening a switch on a live coil throws a spark.\n",{"path":14975,"title":14976,"module":14952,"summary":14977},"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Fmagnetic-energy","Magnetic Energy","Building current in a coil means working against its back emf, and that work does not vanish — it sits in the magnetic field as recoverable energy $U_B=\\tfrac12LI^2$, spread through space at density $u_B=B^2\u002F(2\\mu_0)$. We derive both forms, show they agree for a solenoid, and read a force out of the same energy: an armature is pulled toward higher inductance, and $B^2\u002F(2\\mu_0)$ doubles as a magnetic pressure. The lesson closes on the accounting a real switching event demands, where recoverable energy, copper heating, core loss, and clamp dissipation must balance a single ledger.\n",{"path":14979,"title":14980,"module":14952,"summary":14981},"\u002Felectricity-and-magnetism\u002Felectromagnetic-induction\u002Frl-circuits","RL Circuits","Put a resistor and an inductor in series and the current cannot switch on or off at will: it climbs to $V_0\u002FR$ and falls away exponentially on a single time scale $\\tau=L\u002FR$ set by how much flux the coil hoards against how fast the resistor bleeds it. We solve the turn-on and turn-off, then confront the practical sting — because the coil's current refuses to stop instantly, breaking its path throws up a large voltage, which is why real inductive circuits carry freewheel diodes and clamps that trade voltage stress against how quickly the current dies.\n",{"path":14983,"title":14984,"module":14985,"summary":14986},"\u002Felectricity-and-magnetism\u002Falternating-current\u002Fac-fundamentals","AC Fundamentals","Alternating Current","A wall socket delivers a voltage that averages to zero over each cycle, yet it still heats a filament and runs a motor. The resolution is that dissipation follows the mean of the square, not the mean, so we define the root-mean-square value that makes an alternating source the equal of a DC one for resistive heating. We show a sinusoid's RMS is its peak divided by $\\sqrt2$, work out the average power an ideal resistor draws when its current stays in phase with the applied voltage, and separate the peak, average, and RMS descriptions that a single number cannot combine.\n",{"path":14988,"title":14989,"module":14985,"summary":14990},"\u002Felectricity-and-magnetism\u002Falternating-current\u002Freactance","Reactance","A resistor obeys Ohm's law instant by instant, but a capacitor responds to how fast its voltage changes and an inductor to how fast its current changes. Under a steady sinusoid that rate-dependence collapses to a fixed quarter-cycle phase shift and a frequency-dependent amplitude ratio, the reactance. We derive $X_C=1\u002F(\\omega C)$ and $X_L=\\omega L$, adopt phasors to turn the defining derivatives into multiplication by $j\\omega$ so a single complex impedance carries amplitude and phase together, and track the energy an ideal reactance stores and returns without dissipating it. Real windings and dielectrics add loss, leakage, and self-resonance that bound where the ideal formulas hold.\n",{"path":14992,"title":14993,"module":14985,"summary":14994},"\u002Felectricity-and-magnetism\u002Falternating-current\u002Frlc-resonance","RLC Resonance","Put a resistor, inductor, and capacitor in one loop and their reactances work against each other: inductive reactance grows with frequency while capacitive reactance shrinks, and at one frequency they cancel exactly. There the branch looks purely resistive, the current peaks, and the inductor and capacitor voltages can swing far above the source. We locate that resonance at $\\omega_0=1\u002F\\sqrt{LC}$, measure how sharp the peak is with the quality factor $Q=\\omega_0L\u002FR$, tie its half-power bandwidth $R\u002FL$ to the ringdown of the unforced circuit, and read the same poles off as bandpass and peaked filters at the R, L, or C terminals.\n",{"path":14996,"title":14997,"module":14985,"summary":14998},"\u002Felectricity-and-magnetism\u002Falternating-current\u002Fac-power","AC Power","Multiply an AC load's RMS voltage by its RMS current and you get an answer in volt-amperes that the wiring must carry, but not in general the watts the load consumes. The phase between voltage and current splits that product into a part that does net work and a part that merely sloshes energy back and forth. We derive the average power $P=V_{\\rm rms}I_{\\rm rms}\\cos\\phi$, package amplitude and phase into complex power $S=P+jQ$ so that real, reactive, and apparent power form one right triangle, and see why a harmonic-rich current forces the time-domain definition $P=\\langle vi\\rangle$ in place of a single phase angle.\n",{"path":15000,"title":15001,"module":14985,"summary":15002},"\u002Felectricity-and-magnetism\u002Falternating-current\u002Ftransformers","Transformers","Two coils sharing an iron core exchange no charge, yet a changing current in one drives a voltage in the other, and the ratio of their turns sets how voltage and current trade off between the windings. That lets a transformer step a voltage up or down, isolate two circuits, and make a load look larger or smaller to the source by the square of the turns ratio. We build the ideal ratio element from Faraday's law and the dot convention, derive the reflected-impedance rule, then add the winding resistance, leakage, magnetizing current, and core loss that turn the ideal ratios into real regulation, efficiency, and a bounded voltage-frequency range.\n",{"path":15004,"title":15005,"module":15006,"summary":15007},"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Fdisplacement-current","Displacement Current","Maxwell’s Equations and Electromagnetic Waves","Ampère's law asks for the current through a surface bounded by a loop, but a charging capacitor breaks it: slide the surface off the wire and into the gap and the enclosed conduction current drops to zero, while the magnetic field around the loop plainly does not. Maxwell's repair is to count a changing electric flux as itself a source of magnetic circulation. We derive the displacement-current term $\\varepsilon_0\\,\\d\\Phi_E\u002F\\d t$, show that charge continuity demands it, compute the magnetic field it produces inside a charging capacitor, and see how it closes the Ampère–Maxwell law so that electric and magnetic fields can sustain one another as a wave.\n",{"path":15009,"title":15010,"module":15006,"summary":15011},"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Felectromagnetic-waves","Electromagnetic Waves","Once a changing electric flux can drive a magnetic field, the two curl laws feed each other: a disturbance in one regenerates the other, and the pair walks off through empty space with no medium holding it up. We take the curl of Faraday's law, land on a wave equation whose speed is fixed entirely by $\\mu_0$ and $\\varepsilon_0$, and find that $c=1\u002F\\sqrt{\\mu_0\\varepsilon_0}$ falls out of purely electric and magnetic constants. The plane-wave solution then fixes the geometry — $\\vec E$, $\\vec B$, and the propagation direction mutually perpendicular, oscillating in phase, with amplitudes locked at $E=cB$ — a set of independent predictions any real measurement must meet at once.\n",{"path":15013,"title":15014,"module":15006,"summary":15015},"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Felectromagnetic-momentum","Electromagnetic Momentum","A light beam carries no mass, yet it pushes: shine it on a surface and the surface feels a force. We trace that force back to the fields, which store energy with density $\\varepsilon_0E^2$ and carry it along the Poynting vector $\\vec S=\\vec E\\times\\vec B\u002F\\mu_0$. Because that energy also carries momentum $U\u002Fc$, an absorbed beam presses with $I\u002Fc$ and a mirror with $2I\u002Fc$. We derive the Poynting theorem as local energy conservation, tie intensity to field amplitude, and work the momentum balance carefully enough that oblique incidence, partial reflection, and finite beams all drop out of one accounting.\n",{"path":15017,"title":15018,"module":15006,"summary":15019},"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Fdipole-radiation","Dipole Radiation","Only accelerating charge radiates, and the simplest accelerator is a charge sloshing back and forth: an oscillating electric dipole. We work out the field it throws off, keeping the part that survives to large distance — the $1\u002Fr$ radiation field whose intensity goes as $\\sin^2\\theta\u002Fr^2$, zero along the dipole axis and strongest broadside. From it follow the $\\omega^4$ scaling of total radiated power, radiation resistance as the feed's view of that escaping power, and, through reciprocity, the fact that a good transmitter receives well in the same directions. The near-zone terms that fall off faster carry no net power, and we mark carefully where each description is allowed to be used.\n",{"path":15021,"title":15022,"module":15006,"summary":15023},"\u002Felectricity-and-magnetism\u002Fmaxwell-electromagnetic-waves\u002Fpolarization","Polarization","A plane wave still leaves one thing free: which way its electric field points as it oscillates. That freedom is polarization, set entirely by the relative amplitude and phase of the two transverse field components — in phase gives a line, equal amplitudes a quarter cycle apart give a circle, everything else an ellipse. We work out how a linear analyzer reads a state through Malus's law $I=I_0\\cos^2\\theta$, why that scan alone cannot tell circular light from unpolarized, and how a quarter-wave plate plus a few analyzer settings recover the full Stokes vector and the degree of polarization.\n",{"path":15025,"title":15026,"module":15027,"summary":15028},"\u002Felectricity-and-magnetism\u002Foptics\u002Freflection-and-refraction","Reflection and Refraction","Geometrical Optics","Light meeting a boundary between two transparent media splits into a reflected ray and a bent transmitted one, and predicting where those rays go is the whole starting point of geometrical optics. Fixing one convention — every angle measured from the surface normal — we get reflection's equal angles and derive Snell's law $n_1\\sin\\theta_1=n_2\\sin\\theta_2$ from wavefront timing. That single relation, applied once or twice, yields the critical angle and total internal reflection, prism deviation, the lateral shift through a window, apparent depth, and a fiber's acceptance cone; a wavelength-dependent index then adds dispersion. We mark throughout where the ray picture is trustworthy: feature sizes large against the wavelength and clean interface geometry.\n",{"path":15030,"title":15031,"module":15027,"summary":15032},"\u002Felectricity-and-magnetism\u002Foptics\u002Fthin-lenses","Thin Lenses","A lens gathers the light spreading from one point back onto another, and a single paraxial relation $1\u002Fs+1\u002Fs'=1\u002Ff$ predicts where that image lands and how large it is. We collapse two refractions into one bending plane, read image position and orientation off the three principal rays, and trace focal length back to glass and curvature through the lensmaker equation. Sign conventions carry the physics here — they separate real from virtual images and upright from inverted — so we drill them before chaining lenses in sequence and in contact. The lesson ends on how focal length is actually measured on a bench, and where finite thickness, aperture, and dispersion break the thin-lens picture.\n",{"path":15034,"title":15035,"module":15027,"summary":15036},"\u002Felectricity-and-magnetism\u002Foptics\u002Fspherical-mirrors","Spherical Mirrors","Curve a mirror and it stops merely reflecting an image and starts forming one: the same $1\u002Fs+1\u002Fs'=1\u002Ff$ that governs lenses reappears, now with $f=R\u002F2$ and reflected rays and object sharing one side of the glass. We derive the mirror equation from the reflection geometry of a single paraxial ray, then let signed distances do the sorting — real inverted images on the near branch, virtual upright ones behind the surface — and check the concave, convex, and plane-mirror limits against each other. The second half turns to how focal length is actually measured on a bench, by finite conjugates, distant targets, return imaging, and sagitta, and to the aperture and off-axis aberrations the single paraxial focus cannot capture.\n",{"path":15038,"title":15039,"module":6,"summary":6},"\u002Felectricity-and-magnetism","Electricity & Magnetism",{"path":15041,"title":15042,"module":15043,"summary":15044},"\u002Flinear-algebra\u002Flinear-systems\u002Fsystems-and-echelon-forms","Systems of Linear Equations and Row Reduction","Linear Equations in Linear Algebra","A linear system is a finite set of linear equations in shared variables. Elementary row operations rewrite it without changing its solution set, and reducing the augmented matrix to echelon form decides both existence and uniqueness. Pivot positions say whether the solution set is empty, a single point, or infinite.\n",{"path":15046,"title":15047,"module":15043,"summary":15048},"\u002Flinear-algebra\u002Flinear-systems\u002Fvector-and-matrix-equations","Vector Equations and the Matrix Equation Ax = b","The same linear system reads three equivalent ways: a system of equations, a vector equation asking whether b is a linear combination of fixed vectors, and a matrix equation Ax = b. Ax is the linear combination of A's columns weighted by x, so consistency for a given b means b lies in the span of the columns, and consistency for every b means the columns span all of R^m.\n",{"path":15050,"title":15051,"module":15043,"summary":15052},"\u002Flinear-algebra\u002Flinear-systems\u002Fsolution-sets-and-applications","Solution Sets and Applied Linear Systems","A homogeneous system Ax = 0 has a solution set that is a span through the origin; a consistent Ax = b has that same span translated by any one particular solution. Parametric vector form writes both explicitly. The structure shows up in applied systems with many solutions: equilibrium prices, balanced chemical reactions, network flows, weight-loss diets, and migration models.\n",{"path":15054,"title":15055,"module":15043,"summary":15056},"\u002Flinear-algebra\u002Flinear-systems\u002Flinear-independence","Linear Independence","A set of vectors is linearly independent when the only linear combination equal to zero is the trivial one; otherwise a dependence relation writes one vector in terms of the others. For the columns of A the question becomes whether Ax = 0 has only the trivial solution — a pivot in every column. Counting pivots settles independence, and any set with more vectors than entries is automatically dependent.\n",{"path":15058,"title":15059,"module":15043,"summary":15060},"\u002Flinear-algebra\u002Flinear-systems\u002Flinear-transformations","Linear Transformations and Their Matrices","Reading A as an action rather than an array, x maps to Ax is a transformation from R^n to R^m. The ones that preserve addition and scalar multiplication are the linear transformations, and every one is x maps to Ax for a unique standard matrix whose columns are the images of the standard basis vectors. Onto and one-to-one translate into the span and independence of those columns.\n",{"path":15062,"title":15063,"module":15064,"summary":15065},"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fmatrix-operations","Matrix Operations","Matrix Algebra","Matrices add and scale entrywise, but their product is defined so that multiplication corresponds to composition of linear maps: the columns of AB are A applied to the columns of B. From that requirement follow the row-column rule, the algebra of products (associative and distributive but not commutative), powers, and the transpose.\n",{"path":15067,"title":15068,"module":15064,"summary":15069},"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fmatrix-inverse-and-invertibility","The Inverse and the Invertible Matrix Theorem","The inverse of a square matrix is the matrix analogue of a reciprocal, defined by AA⁻¹ = I. A closed form settles the 2×2 case; the Gauss–Jordan algorithm row reduces [A | I] to [I | A⁻¹] in general; and elementary matrices record single row operations. The Invertible Matrix Theorem collects a dozen equivalent conditions for invertibility into one statement.\n",{"path":15071,"title":15072,"module":15064,"summary":15073},"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fpartitioned-matrices-and-lu","Block Matrices and the LU Factorization","Partitioning a matrix into blocks lets sums, products, and inverses be computed block by block, as if the submatrices were scalars. Block structure also underlies the LU factorization A = LU, which splits solving Ax = b into two fast triangular solves and repays the cost whenever many systems share one coefficient matrix.\n",{"path":15075,"title":15076,"module":15064,"summary":15077},"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fsubspaces-dimension-rank","Subspaces of Rⁿ, Dimension, and Rank","A subspace is a set closed under addition and scalar multiplication. Every matrix carries two: the column space of all attainable outputs Ax, and the null space of all solutions of Ax = 0. A basis measures each with a minimal spanning set, dimension counts it, and the Rank Theorem ties pivots and free variables together as rank + nullity = n.\n",{"path":15079,"title":15080,"module":15064,"summary":15081},"\u002Flinear-algebra\u002Fmatrix-algebra\u002Fapplications-leontief-and-graphics","Applications: Leontief Economics and Computer Graphics","The Leontief input–output model balances an economy through (I − C)x = d and expands the inverse as a geometric series in the consumption matrix. Computer graphics moves figures with matrix products, using homogeneous coordinates so that translation and perspective projection become matrix multiplications too.\n",{"path":15083,"title":15084,"module":15085,"summary":15086},"\u002Flinear-algebra\u002Fdeterminants\u002Fdeterminants-and-cofactors","Introduction to Determinants","Determinants","The determinant of a square matrix is defined recursively by cofactor expansion: an n-by-n determinant is a signed sum of (n-1)-by-(n-1) determinants built from the first row. The expansion can equally run along any row or down any column, and a triangular matrix has determinant equal to the product of its diagonal.\n",{"path":15088,"title":15089,"module":15085,"summary":15090},"\u002Flinear-algebra\u002Fdeterminants\u002Fproperties-of-determinants","Properties of Determinants","Row operations act on the determinant in three predictable ways, and this turns row reduction into a fast algorithm: the determinant is the product of the pivots times a sign for the interchanges. The same properties yield the invertibility test det A is nonzero, the transpose identity, and the multiplicative law det(AB) equals det A times det B.\n",{"path":15092,"title":15093,"module":15085,"summary":15094},"\u002Flinear-algebra\u002Fdeterminants\u002Fcramer-volume-and-area","Cramer's Rule, Volume, and Linear Transformations","Cramer's rule writes each unknown of an invertible system as a ratio of determinants, and the same idea gives a closed formula for the inverse through the adjugate. Geometrically the absolute determinant is the area of the parallelogram or the volume of the parallelepiped spanned by the columns, so a linear map scales every region's measure by that factor.\n",{"path":15096,"title":15097,"module":15098,"summary":15099},"\u002Flinear-algebra\u002Fvector-spaces\u002Fvector-spaces-and-subspaces","Vector Spaces and Subspaces","Vector Spaces","A vector space is any set closed under addition and scalar multiplication that obeys ten algebraic axioms. The same axioms that govern arrows in the plane govern polynomials, functions, matrices, and infinite signals, so one theory covers them all. A subspace is a subset that is a vector space in its own right, tested by three conditions, and the span of any set of vectors is the smallest subspace containing them.\n",{"path":15101,"title":15102,"module":15098,"summary":15103},"\u002Flinear-algebra\u002Fvector-spaces\u002Fnull-and-column-spaces","Null Spaces, Column Spaces, and Linear Transformations","Two subspaces sit inside every matrix. The null space collects all solutions of $Ax = 0$ and lives in the domain; the column space collects every attainable $Ax$ and lives in the codomain. One is defined implicitly by a condition, the other explicitly by a spanning set, and the same pair appears for an abstract linear transformation as its kernel and range.\n",{"path":15105,"title":15106,"module":15098,"summary":15107},"\u002Flinear-algebra\u002Fvector-spaces\u002Fbases-and-independent-sets","Linearly Independent Sets and Bases","A basis is a spanning set with no redundancy: linearly independent and still large enough to reach every vector. The spanning-set theorem shows any spanning set can be trimmed to a basis by discarding dependent vectors, and the pivot columns of a matrix give a basis for its column space. Independence and spanning are defined for abstract spaces exactly as in $\\mathbb{R}^n$.\n",{"path":15109,"title":15110,"module":15098,"summary":15111},"\u002Flinear-algebra\u002Fvector-spaces\u002Fcoordinate-systems","Coordinate Systems","Fixing a basis assigns every vector a unique list of coordinates, turning an abstract space into $\\mathbb{R}^n$. The coordinate mapping is a one-to-one linear transformation onto $\\mathbb{R}^n$ — an isomorphism — so any $n$-dimensional space is indistinguishable from $\\mathbb{R}^n$ as far as vector-space computations go. In $\\mathbb{R}^n$ the change-of-coordinates matrix $P_B$ and its inverse convert between basis coordinates and standard coordinates.\n",{"path":15113,"title":15114,"module":15098,"summary":15115},"\u002Flinear-algebra\u002Fvector-spaces\u002Fdimension-and-rank","The Dimension of a Vector Space and Rank","Every basis of a space has the same number of vectors, and that number is the dimension. Rank is the dimension of the column space, equal to the dimension of the row space and to the number of pivots. The Rank Theorem, rank plus nullity equals the number of columns, ties the four fundamental subspaces of a matrix together and adds six lines to the Invertible Matrix Theorem.\n",{"path":15117,"title":15118,"module":15098,"summary":15119},"\u002Flinear-algebra\u002Fvector-spaces\u002Fchange-of-basis","Change of Basis","Two bases give the same vector two different coordinate vectors, and a single invertible matrix converts between them. Its columns are the coordinate vectors of the old basis expressed in the new one, and its inverse reverses the conversion. In $\\mathbb{R}^n$ the change-of-coordinates matrix between two bases is found by one row reduction.\n",{"path":15121,"title":15122,"module":15098,"summary":15123},"\u002Flinear-algebra\u002Fvector-spaces\u002Fdifference-equations-and-markov","Applications: Difference Equations and Markov Chains","The solutions of an nth-order linear difference equation form an $n$-dimensional vector space, so finding $n$ independent solutions gives them all. A Markov chain evolves a probability distribution by repeated multiplication by a stochastic matrix, and a regular chain converges to a unique steady-state vector fixed by that matrix. Both applications turn a dynamic process into a subspace or a fixed-point question.\n",{"path":15125,"title":15126,"module":15127,"summary":15128},"\u002Flinear-algebra\u002Feigenvalues\u002Feigenvectors-and-eigenvalues","Eigenvectors and Eigenvalues","Eigenvalues and Eigenvectors","An eigenvector of a square matrix is a nonzero vector the matrix only stretches; its eigenvalue is the stretch factor. The eigenspace of an eigenvalue is the null space of A minus lambda times the identity, the eigenvalues of a triangular matrix are its diagonal entries, and eigenvectors for distinct eigenvalues are linearly independent.\n",{"path":15130,"title":15131,"module":15127,"summary":15132},"\u002Flinear-algebra\u002Feigenvalues\u002Fthe-characteristic-equation","The Characteristic Equation","The eigenvalues of a matrix are the roots of its characteristic polynomial det(A minus lambda I). This degree-n polynomial carries an algebraic multiplicity at each repeated root, a nonzero determinant is equivalent to zero not being an eigenvalue, and similar matrices share a characteristic polynomial and hence the same eigenvalues.\n",{"path":15134,"title":15135,"module":15127,"summary":15136},"\u002Flinear-algebra\u002Feigenvalues\u002Fdiagonalization","Diagonalization","A matrix is diagonalizable when it factors as A equals P D P inverse with D diagonal, which happens exactly when it has n linearly independent eigenvectors. The factorization computes matrix powers cheaply, distinct eigenvalues guarantee it, and a repeated eigenvalue permits it only when its eigenspace dimension equals its multiplicity.\n",{"path":15138,"title":15139,"module":15127,"summary":15140},"\u002Flinear-algebra\u002Feigenvalues\u002Feigenvectors-and-linear-transformations","Eigenvectors and Linear Transformations","Every linear transformation between finite-dimensional spaces has a matrix relative to chosen bases, built from the coordinate vectors of the images of the basis vectors. For a map from a space to itself, an eigenvector basis makes that matrix diagonal, and that change of basis is diagonalization; the matrices similar to A are the representations of the map in every basis.\n",{"path":15142,"title":15143,"module":15127,"summary":15144},"\u002Flinear-algebra\u002Feigenvalues\u002Fcomplex-eigenvalues","Complex Eigenvalues","A real matrix with no real eigenvalues still has complex ones, occurring in conjugate pairs. A real 2-by-2 matrix with eigenvalue a plus b i is similar to a rotation-scaling matrix, whose rotation angle is the argument of the eigenvalue and whose scale factor is its modulus; the modulus decides whether the trajectories close up, spiral in, or spiral out.\n",{"path":15146,"title":15147,"module":15127,"summary":15148},"\u002Flinear-algebra\u002Feigenvalues\u002Fdynamical-systems","Discrete and Continuous Dynamical Systems","Eigenvalues govern the long-term behavior of a system that evolves by x becomes A x or by x prime equals A x. An eigenvector basis decouples both kinds of system into independent scalar equations; the eigenvalues then classify the origin as attractor, repeller, saddle, or spiral, and the dominant eigenpair fixes the growth rate and limiting direction.\n",{"path":15150,"title":15151,"module":15127,"summary":15152},"\u002Flinear-algebra\u002Feigenvalues\u002Fpower-method","Iterative Estimates for Eigenvalues","When only a numerical eigenvalue is needed, iteration is preferred over the characteristic polynomial. The power method repeatedly multiplies by A to converge on the dominant eigenvalue and its eigenvector; the Rayleigh quotient sharpens the estimate for symmetric matrices; and the inverse power method targets any eigenvalue near a known guess.\n",{"path":15154,"title":15155,"module":15156,"summary":15157},"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Finner-product-length-orthogonality","Inner Product, Length, and Orthogonality","Orthogonality and Least Squares","The dot product turns the algebra of vectors in R^n into geometry: length, distance, and perpendicularity. The inner product yields the norm, the Pythagorean theorem, and the orthogonal complement, and the null space of a matrix is the orthogonal complement of its row space.\n",{"path":15159,"title":15160,"module":15156,"summary":15161},"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Forthogonal-sets-and-projections","Orthogonal Sets and Orthogonal Projections","An orthogonal basis makes coordinates trivial: each weight is a single dot product, no linear system required. Orthogonal and orthonormal bases give a direct projection formula onto a line and onto a subspace, the orthogonal decomposition and best-approximation theorems, and the matrix form U U-transpose of a projection.\n",{"path":15163,"title":15164,"module":15156,"summary":15165},"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Fgram-schmidt-and-qr","The Gram-Schmidt Process and QR Factorization","Gram-Schmidt turns any basis into an orthogonal one by repeatedly subtracting off projections onto the span already built. Normalizing the result and recording the coefficients factors the matrix as A = QR, with Q orthonormal and R upper triangular, the factorization behind stable least-squares and eigenvalue algorithms.\n",{"path":15167,"title":15168,"module":15156,"summary":15169},"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Fleast-squares-problems","Least-Squares Problems","When Ax = b has no solution, the least-squares solution makes Ax as close to b as possible. The closest Ax is the projection of b onto the column space, and the vector that produces it solves the normal equations A-transpose A x = A-transpose b. Uniqueness, the residual error, and the stabler QR route follow.\n",{"path":15171,"title":15172,"module":15156,"summary":15173},"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Fleast-squares-applications","Applications to Linear Models","Curve fitting is a least-squares problem in statistical notation. The least-squares line, polynomial fits, and multiple regression all reduce to X beta = y with a design matrix X built from the data, solved by the same normal equations.\n",{"path":15175,"title":15176,"module":15156,"summary":15177},"\u002Flinear-algebra\u002Forthogonality-least-squares\u002Finner-product-spaces","Inner Product Spaces","Promoting the four properties of the dot product to axioms defines an inner product on any vector space, including spaces of functions. Length, distance, orthogonality, Gram-Schmidt, and best approximation all carry over, along with the Cauchy-Schwarz and triangle inequalities and the integral inner product behind Fourier approximation.\n",{"path":15179,"title":15180,"module":15181,"summary":15182},"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fdiagonalizing-symmetric-matrices","Diagonalization of Symmetric Matrices","Symmetric Matrices, Quadratic Forms, and the SVD","A symmetric matrix is one that equals its own transpose. Every such matrix can be diagonalized by an orthogonal change of basis, A = PDPᵀ, with real eigenvalues and perpendicular eigenvectors. This is the Spectral Theorem, and it rewrites A as a weighted sum of rank-one projections onto its eigenvectors.\n",{"path":15184,"title":15185,"module":15181,"summary":15186},"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fquadratic-forms","Quadratic Forms","A quadratic form xᵀAx is the second-degree analogue of a linear map, attached to a symmetric matrix A. Orthogonal diagonalization changes variables to the eigenbasis, removing all cross-terms and rotating the form into standard position. The signs of the eigenvalues then classify it as definite or indefinite.\n",{"path":15188,"title":15189,"module":15181,"summary":15190},"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fconstrained-optimization","Constrained Optimization","Maximizing a quadratic form xᵀAx over the unit sphere has an exact answer: the maximum is the largest eigenvalue of A, attained at its eigenvector, and the minimum is the smallest eigenvalue. Adding orthogonality constraints peels off the eigenvalues in order, characterizing the whole spectrum by optimization.\n",{"path":15192,"title":15193,"module":15181,"summary":15194},"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fsingular-value-decomposition","The Singular Value Decomposition","The singular value decomposition factors any m×n matrix as A = UΣVᵀ, with orthogonal U and V and a nonnegative diagonal Σ of singular values. The singular values are the square roots of the eigenvalues of AᵀA, and they describe the matrix geometrically as a rotation, an axiswise stretch, and another rotation, exposing rank, the four fundamental subspaces, and a best low-rank approximation.\n",{"path":15196,"title":15197,"module":15181,"summary":15198},"\u002Flinear-algebra\u002Fsymmetric-quadratic-svd\u002Fsvd-applications-pca-imaging","Applications: Image Processing and Statistics","Principal component analysis diagonalizes the covariance matrix of a data set, producing uncorrelated variables ordered by variance. The leading components capture most of the variation, which reduces dimension, compresses images through low-rank SVD approximation, and connects directly to the singular values of the data matrix.\n",{"path":15200,"title":15201,"module":15202,"summary":15203},"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fnumerical-thinking-and-matrix-computation","Numerical Thinking and Matrix Computation","Numerical Linear Algebra","Numerical analysis builds efficient discrete algorithms for continuous problems, and its cost is dominated as much by memory traffic as by arithmetic. Block matrix calculus, flop counts, and the BLAS efficiency ratio fix the cost model; triangular and unitary matrices are the two computational building blocks every factorization rests on.\n",{"path":15205,"title":15206,"module":15202,"summary":15207},"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Flu-and-cholesky","LU and Cholesky Factorization in Practice","Gaussian elimination, read as a factorization A = LU, turns a linear system into two triangular solves. A single near-zero pivot wrecks it, so partial pivoting reorders rows to pick the largest available pivot and makes the method work for every invertible matrix. For symmetric positive-definite systems, Cholesky halves the cost and needs no pivoting.\n",{"path":15209,"title":15210,"module":15202,"summary":15211},"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fconditioning-and-floating-point","Conditioning and Floating-Point Arithmetic","A problem's condition number measures how much its answer moves when its data is perturbed, independent of any algorithm. Subtraction is ill-conditioned under cancellation, and for a linear system the amplifier is the matrix condition number κ(A). Floating-point arithmetic supplies the perturbation: every real number is rounded to within a relative machine precision, so even perfect computation inherits an error of order κ times the unit roundoff.\n",{"path":15213,"title":15214,"module":15202,"summary":15215},"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fstability-and-error-analysis","Numerical Stability and Backward Error Analysis","An algorithm is backward stable when its computed answer is the exact answer to a slightly perturbed problem. Combined with the condition number this gives the governing rule of thumb: forward error is at most condition times stability. Three cancellation case studies make the point, then the residual-based backward error applies it to Ax = b and shows why partial pivoting keeps Gaussian elimination stable.\n",{"path":15217,"title":15218,"module":15202,"summary":15219},"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fqr-and-numerical-least-squares","QR, Householder, and Numerical Least Squares","The least-squares problem reduces to the normal equations, but forming AᵀA squares the condition number and can wreck accuracy. The stable route computes a QR factorization directly on A and solves Rx = Qᵀb. Householder reflectors build that QR one column at a time using length-preserving reflections, the unconditionally backward-stable building block behind every serious least-squares solver.\n",{"path":15221,"title":15222,"module":15202,"summary":15223},"\u002Flinear-algebra\u002Fnumerical-linear-algebra\u002Fnumerical-eigenvalues-and-svd","Numerical Eigenvalue Problems and the SVD","Eigenvalues cannot be found by a formula for large matrices, so they are found by iteration. Power and inverse iteration converge to one eigenvector at a rate set by the eigenvalue gap; the QR algorithm sweeps a matrix to Schur form and, with a good shift and a Hessenberg reduction, computes the whole spectrum in cubic time. Singular values follow from the same machinery applied without ever forming AᵀA.\n",{"path":15225,"title":15226,"module":15227,"summary":15228},"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Faffine-combinations","Affine Combinations","Geometry of Vector Spaces","An affine combination is a linear combination whose weights sum to one. The affine hull of a set is the smallest flat containing it: a point, a line, a plane, or a translated subspace. Homogeneous coordinates turn every affine combination into an ordinary linear combination one dimension up.\n",{"path":15230,"title":15231,"module":15227,"summary":15232},"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Faffine-independence-and-barycentric-coordinates","Affine Independence and Barycentric Coordinates","Affine independence is linear independence for the translated or lifted points, and it guarantees each point of an affine hull a unique weight vector. Those weights are barycentric coordinates: centers of mass, ratios of triangle areas, and the interpolation rule behind smooth shading in computer graphics.\n",{"path":15234,"title":15235,"module":15227,"summary":15236},"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Fconvex-combinations-and-convex-sets","Convex Combinations and Convex Sets","A convex combination is an affine combination with nonnegative weights, and the convex hull of a set is the smallest convex set containing it. Convex sets are closed under intersection, and Carathéodory's theorem bounds how many points a convex combination in $\\mathbb{R}^n$ ever needs: at most $n+1$.\n",{"path":15238,"title":15239,"module":15227,"summary":15240},"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Fhyperplanes-and-polytopes","Hyperplanes and Polytopes","A hyperplane is a level set of a linear functional, the set where an inner product equals a constant. Hyperplanes separate disjoint convex sets and support them at their boundaries. Polytopes are convex hulls of finite point sets; their vertices are the extreme points, and a linear functional attains its extremes there.\n",{"path":15242,"title":15243,"module":15227,"summary":15244},"\u002Flinear-algebra\u002Fgeometry-of-vector-spaces\u002Fcurves-and-surfaces","Curves and Surfaces","Bézier curves are affine combinations of control points with polynomial weights, so they lie in the convex hull of those points and bend toward them. The de Casteljau algorithm evaluates them by repeated interpolation, a matrix form factors them for computation, and matching endpoints and tangents joins segments into smooth curves and surfaces.\n",{"path":15246,"title":15247,"module":6,"summary":6},"\u002Flinear-algebra","Linear Algebra",{"path":15249,"title":15250,"module":6,"summary":6},"\u002Ftheory-of-computation","Theory of Computation",{"path":15252,"title":15253,"module":14054,"summary":15254},"\u002Fcomputer-architecture\u002Ffoundations\u002Fbits-bytes-and-words","Bits, Bytes, and Words","Everything a machine stores is a string of bits grouped into bytes. We set out binary and hexadecimal, the byte as the unit of addressing, the word as the machine's natural integer size, and byte ordering — why the same four bytes read as 0x01234567 on one machine and 0x67452301 on another.\n",{"path":15256,"title":15257,"module":14054,"summary":15258},"\u002Fcomputer-architecture\u002Ffoundations\u002Finteger-representation","Integer Representation","A fixed-width byte string is just a pattern; what makes it a number is the rule we read it by. We define unsigned encoding and two's complement — where the top bit carries a negative weight — derive the ranges UMax, TMin, and TMax, and show how the same bits reinterpret between signed and unsigned, how widening sign-extends, and what truncation throws away.\n",{"path":15260,"title":15261,"module":14054,"summary":15262},"\u002Fcomputer-architecture\u002Ffoundations\u002Finteger-arithmetic","Integer Arithmetic","Fixed-width integer arithmetic is arithmetic modulo a power of two: add past the top and the result wraps. We work out unsigned and two's-complement addition and the rules that detect their overflow, why negation is a complement-plus-one, how multiplication truncates to the low-order bits and how compilers turn constant multiplies into shifts and adds, why C declares signed overflow undefined, and the bias fix that keeps shift-based signed division rounding toward zero.\n",{"path":15264,"title":15265,"module":14054,"summary":15266},"\u002Fcomputer-architecture\u002Ffoundations\u002Ffloating-point","Floating Point","IEEE-754 trades the exactness of integers for enormous range by storing numbers as sign, exponent, and fraction — scientific notation in binary. We lay out the single and double formats, the bias that encodes the exponent, the three regimes (normalized, denormalized, special), a worked encode\u002Fdecode, the four rounding modes and round-to-even at the bit level, why addition is not associative, the pitfalls of float-int conversion, and why 0.1 has no exact binary representation.\n",{"path":15268,"title":15269,"module":14054,"summary":15270},"\u002Fcomputer-architecture\u002Ffoundations\u002Fboolean-algebra-and-bit-manipulation","Boolean Algebra and Bit Manipulation","Treat a word as a vector of independent bits and the bitwise operators become an algebra. We define AND, OR, NOT, and XOR as bit vectors, build the masking idioms that set, clear, toggle, and test individual bits, extract fields with zero- and sign-extension, count set bits three ways, derive the classic x & (x - 1) family of tricks, and distinguish bitwise operators from C's short-circuiting logical operators.\n",{"path":15272,"title":15273,"module":15274,"summary":15275},"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fthe-machines-view","The Machine's View","Machine-Level Programming","The instruction set architecture is the contract a compiler writes against: the program counter, sixteen integer registers with their sub-register widths, and the condition codes. We follow one C function down through gcc to assembly, learn to read an instruction as operation plus operands, and fix the vocabulary the rest of the module uses.\n",{"path":15277,"title":15278,"module":15274,"summary":15279},"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fdata-movement","Data Movement","Most instructions a program runs simply move data. We cover the mov family and its size suffixes, the three operand forms, the full memory addressing mode D(Rb,Ri,S) and its special cases, lea for address arithmetic, and how push and pop manipulate the stack pointer %rsp on a stack that grows toward lower addresses.\n",{"path":15281,"title":15282,"module":15274,"summary":15283},"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Farithmetic-and-logic","Arithmetic and Logic","The ALU instructions that compute on register and memory values: add, sub, and imul; the unary inc\u002Fdec\u002Fneg\u002Fnot; the shifts sal\u002Fshr\u002Fsar; the bitwise and\u002For\u002Fxor; and lea reused as a fast arithmetic trick. Each binary operation also sets the condition-code flags CF, ZF, SF, and OF, which cmp and test compute without keeping a result.\n",{"path":15285,"title":15286,"module":15274,"summary":15287},"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fcontrol-flow","Control Flow","How a flat instruction stream realizes branches and loops. The conditional jumps read the condition-code flags; set instructions turn flags into a 0\u002F1 byte. We translate if\u002Felse into the standard compare-and-branch pattern, while\u002Ffor loops into the guarded-do form, and dense switches into jump tables that index a target directly.\n",{"path":15289,"title":15290,"module":15274,"summary":15291},"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fprocedures","Procedures","How a function call works at the machine level: the run-time stack, call and ret passing control through a saved return address, the System V convention that routes the first six arguments through %rdi..%r9 and the result through %rax, the caller-saved versus callee-saved split, the stack frame, and a recursive factorial traced through its frames.\n",{"path":15293,"title":15294,"module":15274,"summary":15295},"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Farrays-structs-and-alignment","Arrays, Structs, and Alignment","How aggregate data lays out in memory. Arrays as base-plus-scaled-index, the row-major ordering of multidimensional arrays, pointer arithmetic in units of the pointed-to type, struct fields at fixed byte offsets, the overlapping storage of unions, and the alignment rules that force padding into a struct.\n",{"path":15297,"title":15298,"module":15274,"summary":15299},"\u002Fcomputer-architecture\u002Fmachine-level-x86-64\u002Fmemory-layout-and-buffer-overflows","Memory Layout and Buffer Overflows","The process address space — text, data, heap, and stack — and the classic vulnerability it enables. A stack buffer that is written past its end can overwrite the saved return address and redirect ret, so we sketch the mechanism defensively and then the three standard protections: stack canaries, a non-executable stack, and address-space layout randomization.\n",{"path":15301,"title":15302,"module":15303,"summary":15304},"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Fwhat-an-isa-is","What an ISA Is","Instruction Set Architecture","The instruction set architecture is the contract that lets a compiler and a chip be written by people who never meet: the stable interface software targets and hardware implements. We separate architecture from microarchitecture, read RISC and CISC as opposite answers to where complexity should live, price out what each choice costs in decode hardware, code density, and pipeline friendliness, and see how x86-64 endures by translating its instructions into RISC-like operations on the fly.\n",{"path":15306,"title":15307,"module":15303,"summary":15308},"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Finstruction-formats-and-operands","Instruction Formats and Operands","An instruction is an opcode plus a way to name its operands. We count operands — 3-address, 2-address, 1-address accumulator, and 0-address stack machines — by writing the same C = A + B four ways, weigh register operands against memory operands, then lay out the same add byte by byte in x86-64 (REX prefix, opcode, ModRM) and in Y86-64, and what fixed versus variable length costs at fetch time.\n",{"path":15310,"title":15311,"module":15303,"summary":15312},"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Faddressing-modes","Addressing Modes","Once an operand field exists, it needs a rule for turning its bits into the data it names. That rule is the addressing mode. We walk the standard set — immediate, register, direct, register-indirect, displacement, scaled-indexed, and PC-relative — fixing the effective-address computation for each, run every mode against one concrete machine state, and price out what Y86-64 loses by keeping only base plus displacement.\n",{"path":15314,"title":15315,"module":15303,"summary":15316},"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Fthe-y86-64-instruction-set","The Y86-64 Instruction Set","Y86-64 is a teaching ISA — a stripped-down x86-64 simple enough to implement by hand yet real enough to compile to. We fix its programmer-visible state (fifteen registers, three condition codes, the PC, memory, and a status code), give the instruction set with exact byte encodings, spell out how the condition codes decide every jXX and cmovXX, and run the encoding both directions: assembly to bytes and raw bytes back to meaning.\n",{"path":15318,"title":15319,"module":15303,"summary":15320},"\u002Fcomputer-architecture\u002Finstruction-set-architecture\u002Fy86-64-programming","Y86-64 Programming","With the encodings fixed, we write real Y86-64 assembly: the .pos, .align, and .quad directives, the calling convention borrowed from x86-64, a stack set up by hand, and complete programs — an array sum and a branch-free max. We watch the assembler turn the listing into the exact byte image the processor will execute, and trace the stack across the call.\n",{"path":15322,"title":15323,"module":15324,"summary":15325},"\u002Fcomputer-architecture\u002Fdigital-logic\u002Ftransistors-gates-and-boolean-functions","Transistors, Gates, and Boolean Functions","Digital Logic","A processor is built from millions of transistor switches. We start at the MOS transistor as a voltage-controlled switch, build the CMOS inverter and NAND transistor by transistor, meet the seven standard gates with their truth tables, show that NAND alone is functionally complete, price each gate in transistors and in time, and turn any truth table into a sum-of-products circuit.\n",{"path":15327,"title":15328,"module":15324,"summary":15329},"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fcombinational-logic-and-hcl","Combinational Logic and HCL","A combinational circuit is a pure Boolean function of its current inputs — no memory, no clock. We draw the line between combinational and sequential logic, do the gate-delay accounting that finds a circuit's critical path and bounds the clock, meet don't-cares, then introduce CS:APP's Hardware Control Language: bit-level operators, word-level signals, equality nets, and the case expression that compiles to a multiplexer tree.\n",{"path":15331,"title":15332,"module":15324,"summary":15333},"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fmultiplexers-decoders-and-the-alu","Multiplexers, Decoders, and the ALU","The combinational building blocks that make a datapath. We build the 2:1 and 4:1 multiplexer and tie it back to HCL's case expression, the n-to-2^n decoder, a one-bit full adder (sum is XOR, carry is majority), the ripple-carry adder that chains them, and finally the ALU — a function unit that selects among add, sub, and, and xor under a control input and exposes condition flags.\n",{"path":15335,"title":15336,"module":15324,"summary":15337},"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fmemory-elements-latches-flip-flops-and-clocking","Memory Elements: Latches, Flip-Flops, and Clocking","A combinational circuit holds no state; feeding a circuit's output back to its input creates memory. We build the SR latch from cross-coupled gates, the level-sensitive D latch, and the master\u002Fslave edge-triggered D flip-flop, then introduce the clock and the synchronous design discipline, the setup\u002Fhold timing window, clock skew, metastability, and the register as n flip-flops sharing one clock.\n",{"path":15339,"title":15340,"module":15324,"summary":15341},"\u002Fcomputer-architecture\u002Fdigital-logic\u002Fregister-files-and-random-access-memory","Register Files and Random-Access Memory","Storage organized for access by address. We build the register file (a small bank of registers with addressed read ports and clocked write ports, the exact structure Y86-64's decode and write-back stages use), then descend to the SRAM and DRAM cells of main memory, why one is fast and dear and the other dense and slow, and how a row decoder picks a word out of a memory array.\n",{"path":15343,"title":15344,"module":15345,"summary":15346},"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fthe-fetch-decode-execute-cycle","The Fetch-Decode-Execute Cycle","Processor Design","A processor is a machine that repeats one loop forever: read the next instruction from memory, figure out what it asks for, do it, and advance. We fix the stored-program idea, lay out the datapath at a high level — PC, instruction memory, register file, ALU, data memory — and the control unit that sequences them, break the work into the six stages the rest of the module builds in hardware, and work out exactly how fetch parses variable-length instructions and computes the next PC.\n",{"path":15348,"title":15349,"module":15345,"summary":15350},"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fthe-seq-stages","The SEQ Stages","The six SEQ stages, made exact. For every Y86-64 instruction — halt, nop, the moves, OPq, the jumps, call and ret, pushq and popq — we write down what Fetch, Decode, Execute, Memory, Write-back, and PC update each compute, as per-instruction stage tables with every row justified. Once the tables are filled in, the processor is fully specified; the remaining lessons turn them into wires.\n",{"path":15352,"title":15353,"module":15345,"summary":15354},"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fcontrol-logic-and-sequencing","Control Logic and Sequencing","The stage tables say what each instruction needs; the control logic computes it from icode. We write the HCL for the register-port selections (srcA, srcB, dstE, dstM), the ALU function and input selection, the memory read\u002Fwrite and address, the branch condition, and the next-PC mux — each a case expression on icode that compiles to a mux — and see how one blob of combinational logic serves every instruction at once. We close by contrasting hardwired control with the microprogrammed alternative.\n",{"path":15356,"title":15357,"module":15345,"summary":15358},"\u002Fcomputer-architecture\u002Fprocessor-design\u002Fassembling-seq","Assembling SEQ","We wire the whole thing together. The functional units from digital logic and the control signals from the last lesson assemble into the complete SEQ datapath, laid out the way CS:APP draws it — six stages stacked bottom to top, Fetch at the floor and PC update at the ceiling, signals flowing up the margins. Then the timing analysis: why everything must settle in one cycle, the no-reading-back principle that makes single-cycle execution consistent, and the critical path that sets the clock. We close by walking an OPq and a ret through the assembled machine.\n",{"path":15360,"title":15361,"module":15345,"summary":15362},"\u002Fcomputer-architecture\u002Fprocessor-design\u002Ftracing-a-program","Tracing a Program","To close the module, we take a complete Y86-64 program — a loop that sums 1 through 3 — and run it through SEQ one cycle at a time, recording the PC, the fetched instruction, every stage computation, and the registers, condition codes, and memory after each cycle. Then we examine single cycles in detail: every named signal of an OPq in concrete hex, and a second program whose call and ret we trace through the stack. The traces confirm that the assembled datapath and control logic behave as a processor.\n",{"path":15364,"title":15365,"module":15366,"summary":15367},"\u002Fcomputer-architecture\u002Fpipelining\u002Fpipelining-principles","Pipelining Principles","Pipelining","A processor that runs one instruction to completion before starting the next wastes most of its hardware most of the time. Pipelining splits the work into stages separated by registers so several instructions are in flight at once. We separate throughput from latency, work the 300 ps example through one, two, and three stages, and derive the three ceilings on the gain: uneven stages, register overhead, and the dependencies between instructions.\n",{"path":15369,"title":15370,"module":15366,"summary":15371},"\u002Fcomputer-architecture\u002Fpipelining\u002Ffrom-seq-to-pipe","From SEQ to PIPE","We turn the sequential Y86-64 processor into a pipelined one by inserting pipeline registers between its stages so each cycle holds one instruction per stage. Doing it correctly forces a rearrangement: the next-PC computation must move into Fetch as a prediction, because the later stages that used to compute it are now busy with other instructions. We walk SEQ to SEQ+ to PIPE, spell out exactly what each pipeline register carries, and fix the naming discipline (D_stat versus d_stat) that keeps five in-flight instructions straight.\n",{"path":15373,"title":15374,"module":15366,"summary":15375},"\u002Fcomputer-architecture\u002Fpipelining\u002Fdata-hazards-stalling-and-forwarding","Data Hazards: Stalling and Forwarding","Overlapping instructions collide when a later one needs a value an earlier one has not finished computing: a read-after-write data hazard. We map exactly which instruction distances are dangerous, fix hazards the slow way by stalling (three bubbles), then the fast way by forwarding from five distinct sources into Decode, in a priority order that sequential semantics forces. Forwarding handles almost everything; the load-use hazard still needs exactly one stall.\n",{"path":15377,"title":15378,"module":15366,"summary":15379},"\u002Fcomputer-architecture\u002Fpipelining\u002Fcontrol-hazards-and-branch-prediction","Control Hazards and Branch Prediction","A pipeline must fetch an instruction every cycle, but after a conditional jump or a ret the next address is not yet known: a control hazard. We measure the branch penalty, weigh predict-taken against its alternatives with real loop arithmetic, watch PIPE detect a misprediction in Execute and squash the two wrong-path instructions, and meet the ret hazard, which has nothing to predict and stalls three cycles. A 2-bit counter gives a taste of dynamic prediction.\n",{"path":15381,"title":15382,"module":15366,"summary":15383},"\u002Fcomputer-architecture\u002Fpipelining\u002Fthe-complete-pipe-processor","The Complete PIPE Processor","We assemble the full pipelined Y86-64: five stages, five pipeline registers, forwarding paths, and a small control unit that decides, each cycle, whether to stall or bubble each register. The subtle part is when hazards combine: one pairing hides a genuine bug. A fourth control case reads stat and keeps exceptions precise. Performance reduces to CPI = 1 + lp + mp + rp, worked out to 1.27 with realistic frequencies, and PIPE beats SEQ by several times despite every penalty.\n",{"path":15385,"title":15386,"module":15387,"summary":15388},"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fstorage-technologies-and-the-latency-gap","Storage Technologies and the Latency Gap","The Memory Hierarchy","No single memory is both fast and large and cheap. We survey the technologies a machine can store bits in — SRAM, DRAM, flash, and rotating disk — open up a DRAM chip to find the row buffer, work a disk access down to the millisecond, and rank everything by speed, density, and cost per bit. Then we watch the processor outrun memory decade after decade. That widening gap is the whole reason a machine stacks fast small storage on top of slow large storage into a hierarchy.\n",{"path":15390,"title":15391,"module":15387,"summary":15392},"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Flocality","Locality","A hierarchy only pays off because programs do not touch memory at random. They reuse recently-used data (temporal locality) and touch nearby data soon after (spatial locality). We make both precise and then quantitative: miss rates for stride-1 and stride-k traversals against a concrete block size, and the loop-order pair on a 2-D array where the same sum misses 16 times one way and 64 times the other — why row-major versus column-major order can change a program's speed by an order of magnitude.\n",{"path":15394,"title":15395,"module":15387,"summary":15396},"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fcache-memories-direct-mapped","Cache Memories and Direct Mapping","A cache is fast SRAM that holds copies of recently-used blocks of main memory. We fix its organization — S sets, E lines per set, B bytes per block — and the way it dissects an address into tag, set index, and block offset, worked bit by bit on a concrete 16-byte cache. Then we run the direct-mapped (E=1) access algorithm end to end on a seven-access trace: index to a set, compare the tag, hit or miss, evict. Cold and conflict misses fall out of the structure, and a two-array ping-pong shows conflict thrashing and its padding fix.\n",{"path":15398,"title":15399,"module":15387,"summary":15400},"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fset-associative-and-write-policies","Set-Associative Caches and Write Policies","Give each set several lines and a block has a choice of homes — fewer conflict misses, at the cost of comparing E tags in parallel and choosing a victim to evict. We re-run the direct-mapped ping-pong trace on a 2-way cache and watch the conflicts vanish, weigh LRU against random replacement, then turn to writes: write-through versus write-back with a dirty bit on a hit, write-allocate versus no-write-allocate on a miss, and a worked traffic count showing when each pairing wins.\n",{"path":15402,"title":15403,"module":15387,"summary":15404},"\u002Fcomputer-architecture\u002Fmemory-hierarchy\u002Fcache-performance-and-cache-friendly-code","Cache Performance and Cache-Friendly Code","Turn the cache mechanism into a number. Hit time, miss rate, and miss penalty combine into the average memory access time; we compute AMAT for a two-level hierarchy with real numbers, weigh the design knobs against each other, and read the memory mountain. Then we write cache-friendly code — the matrix-multiply loop-order case study (ijk versus kij, misses counted per iteration) and loop blocking, where cache-sized tiles turn evicted reuse back into hits.\n",{"path":15406,"title":15407,"module":15408,"summary":15409},"\u002Fcomputer-architecture\u002Fvirtual-memory\u002Faddress-spaces-and-translation","Address Spaces and Translation","Virtual Memory","Every process runs as if it owns a private, contiguous span of memory — its virtual address space — while the hardware maps those addresses onto a single shared physical memory. We fix virtual memory's three jobs (a cache for disk, a memory manager, a protection boundary), the page as the unit of mapping, and the MMU replacing the virtual page number while the offset passes through untouched — then run one translation end to end at the bit level and trace the control flow of a page hit against a page fault.\n",{"path":15411,"title":15412,"module":15408,"summary":15413},"\u002Fcomputer-architecture\u002Fvirtual-memory\u002Fpage-tables-and-page-faults","Page Tables and Page Faults","The page table is an array of page-table entries indexed by virtual page number; each entry's valid bit says whether the page is in DRAM, on disk, or unallocated, and its permission, reference, and dirty bits drive protection and replacement. We walk translation as a table lookup, the page fault and demand paging, the clock algorithm the OS uses to approximate LRU, memory mapping and copy-on-write (why fork is cheap), the taxonomy of bad references, and thrashing.\n",{"path":15415,"title":15416,"module":15408,"summary":15417},"\u002Fcomputer-architecture\u002Fvirtual-memory\u002Fthe-tlb-and-multi-level-page-tables","The TLB and Multi-Level Page Tables","A page-table read on every access would double memory traffic; a flat table for a 48-bit space would occupy 512 GB per process. The TLB fixes the first: a small set-associative cache of PTEs inside the MMU whose tag and index come from the VPN. Multi-level page tables fix the second, allocating only the sub-tables a process uses; x86-64 walks four levels with a 9+9+9+9+12 split. We trace one reference end to end through TLB, walk, and cache, and close with the overlap trick that lets the L1 cache start before translation ends.\n",{"path":15419,"title":15420,"module":15421,"summary":15422},"\u002Fcomputer-architecture\u002Fexceptions-and-io\u002Fexceptional-control-flow","Exceptional Control Flow","Exceptions & I\u002FO","Beyond the sequential, branch, and call flow a program controls itself, the hardware can divert the processor in response to events. We sort these into four classes — interrupts (asynchronous, from devices), traps (intentional syscalls), faults (recoverable, like a page fault), and aborts (unrecoverable) — then take the mechanism apart: exception numbers and the table dispatch, what the hardware pushes and why it differs from a procedure call, the divide-error \u002F page-fault \u002F general-protection trio on x86-64, the full syscall round trip with a worked write in assembly, and processes and signals as the abstractions ECF makes possible.\n",{"path":15424,"title":15425,"module":15421,"summary":15426},"\u002Fcomputer-architecture\u002Fexceptions-and-io\u002Finterrupts-and-the-kernel","Interrupts and the Kernel","An I\u002FO device signals completion by raising an interrupt, crossing the privilege boundary from user mode into the kernel. We fix that boundary, follow an interrupt from device through the interrupt controller to its vectored handler, and use the timer interrupt to drive preemptive scheduling and the context switch. Then the I\u002FO mechanics: polling versus interrupt-driven I\u002FO with a cycle count, device registers and memory-mapped I\u002FO versus port I\u002FO, DMA's full transfer walkthrough and its cache hazard, and a disk read traced end to end, from the read syscall to the completion interrupt.\n",{"path":15428,"title":15429,"module":15430,"summary":15431},"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fprocesses-threads-and-parallelism","Processes, Threads, and Parallelism","Multithreading & Multicore","Around 2004 the single core stopped getting faster, and the industry's answer was to hand programmers more cores instead. This lesson builds the vocabulary that shift demands: process versus thread and exactly which hardware state each one owns, concurrency versus parallelism, the three kinds of parallelism a machine can exploit, why Dennard scaling ended and forced the multicore turn, and Amdahl's law — the arithmetic that bounds the speedup those cores can deliver.\n",{"path":15433,"title":15434,"module":15430,"summary":15435},"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fhardware-multithreading","Hardware Multithreading","A pipeline spends much of its life waiting — on cache misses, on dependences, on branches. Hardware multithreading fills the dead cycles with instructions from another thread. We compare coarse-grained switching (change threads on a long stall), fine-grained interleaving (change every cycle), and simultaneous multithreading (mix threads inside a single cycle), work out exactly which hardware a second thread context duplicates and which it shares, and weigh when SMT pays off and when two threads just fight over one cache.\n",{"path":15437,"title":15438,"module":15430,"summary":15439},"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fcache-coherence","Cache Coherence","Give each core its own cache and the same address can live in two places at once, with copies that disagree. We reproduce the stale-copy bug with a two-core trace, then fix it the way hardware does: snooping caches that watch a shared bus and keep every line in a protocol state. We build MSI in full, upgrade it to MESI, contrast invalidation with updating, add coherence misses as the fourth C, and end with false sharing: the performance bug where cores fight over a line while never touching the same byte.\n",{"path":15441,"title":15442,"module":15430,"summary":15443},"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fmemory-consistency-and-synchronization","Memory Consistency and Synchronization","Coherence keeps cores agreeing about one location; consistency is the contract about many. We define sequential consistency, then watch real hardware break it: the store buffer lets a load slip ahead of an older store, and the classic two-thread litmus test ends with both sides reading zero. We state x86-TSO precisely, restore order with mfence, build atomic read-modify-write from the lock prefix, xchg, and cmpxchg, and write a spinlock twice — once naively, once bus-friendly — closing with what lock-free progress actually guarantees.\n",{"path":15445,"title":15446,"module":15430,"summary":15447},"\u002Fcomputer-architecture\u002Fmultithreading-and-multicore\u002Fmulticore-organization","Multicore Organization","Where everything sits on the die. A modern die gives each core private L1 and L2 caches, spreads a shared last-level cache across slices, and wires it all together with a ring or mesh; multi-socket servers add NUMA, where memory is local to one socket and every remote access pays a latency penalty. We walk the floorplan, put numbers on local versus remote latency, meet thread affinity, and account for the two shared resources — coherence traffic and LLC capacity — that decide how far a parallel program scales.\n",{"path":15449,"title":15450,"module":15451,"summary":15452},"\u002Fcomputer-architecture\u002Fcapstone\u002Fthe-whole-machine","The Whole Machine","Capstone","We take one line of C down the whole tower the course built — compiler to assembly, assembly to machine-code bytes, the bytes into the fetch–decode–execute datapath — then trace one load and one add through the pipelined, cached, translated, interruptible machine, each step cross-linked to the lesson that built it. We close with the map of the course as a stack of layers and an accounting of what we simplified: out-of-order execution, superscalar issue, and speculation past the branch predictor.\n",{"path":15454,"title":15455,"module":15451,"summary":15456},"\u002Fcomputer-architecture\u002Fcapstone\u002Fassembling-a-complete-cpu","Assembling a Complete CPU","We bolt the parts the course built — PC, instruction memory and its fetch logic, register file, ALU, condition codes, data memory, and the control unit — into one complete CPU, name the lesson that built each, wire them in a deliberate order, and power the machine on from reset. Then we assemble a real test program (sum a four-element array through a call\u002Fret procedure), give its exact bytes and memory layout, and trace it cycle by cycle to the answer 0xabcdabcdabcd. We close with how to validate such a machine, and what it takes to put two of them on one die.\n",{"path":15458,"title":15459,"module":6,"summary":6},"\u002Fcomputer-architecture","Computer Architecture",{"path":15461,"title":15462,"module":14054,"summary":15463},"\u002Fdifferential-equations\u002Ffoundations\u002Fmodels-and-direction-fields","Models, Direction Fields, and Solution Curves","A differential equation relates an unknown function to its own rates of change. Three first-order models — a falling body, a cooling object, a population under predation — share the form dy\u002Fdt = ay - b; the slope field fixes their equilibria and long-run behavior before any formula is found. Solving the linear case gives the general solution, its integral curves, and the particular solution selected by an initial condition.\n",{"path":15465,"title":15466,"module":14054,"summary":15467},"\u002Fdifferential-equations\u002Ffoundations\u002Fclassification-and-terminology","Classifying Equations: Order, Linearity, ODE vs. PDE","Every solution method targets a specific class of equation, so the first question about any differential equation is which classes it belongs to. Four independent axes sort them: ordinary versus partial, order, linear versus nonlinear, and homogeneous versus nonhomogeneous. Systems, verification of a solution by substitution, and the split between initial and boundary value problems complete the vocabulary.\n",{"path":15469,"title":15470,"module":15471,"summary":15472},"\u002Fdifferential-equations\u002Ffirst-order\u002Flinear-first-order-integrating-factors","Linear Equations and Integrating Factors","First-Order Equations","A first-order linear equation has the unknown and its derivative to the first power only. Multiplying by an integrating factor collapses the left side into a single derivative, and one integration gives the general solution in closed form. The solution exists wherever the coefficients are continuous, and for a constant coefficient it splits into a decaying transient and a steady state set by the forcing.\n",{"path":15474,"title":15475,"module":15471,"summary":15476},"\u002Fdifferential-equations\u002Ffirst-order\u002Fseparable-and-exact","Separable and Exact Equations","Two nonlinear first-order classes solve by direct integration. A separable equation splits so that each variable can be integrated on its own side, giving an implicit relation. An exact equation is the total differential of a hidden potential function, recognized by a symmetry test on its coefficients; when the test fails, an integrating factor can sometimes restore exactness. A change of variable brings homogeneous equations into the separable class.\n",{"path":15478,"title":15479,"module":15471,"summary":15480},"\u002Fdifferential-equations\u002Ffirst-order\u002Fmodeling-first-order","Modeling with First-Order Equations","A rate law is a differential equation. Each first-order model starts from one governing principle: conservation of mass for a mixing tank, proportional change for interest and radioactive decay, Newton's law of cooling, a force balance for a body falling against drag, and Kirchhoff's law for a series circuit. Setting the derivative to zero recovers the steady state, and the transient records how the initial condition relaxes toward it.\n",{"path":15482,"title":15483,"module":15471,"summary":15484},"\u002Fdifferential-equations\u002Ffirst-order\u002Fautonomous-and-population-dynamics","Autonomous Equations, Phase Lines, and Population Dynamics","An autonomous equation y' = f(y) can be analyzed qualitatively without being solved. Its constant solutions are the zeros of f, and the sign of f between them fixes whether nearby solutions rise or fall, which the phase line records as a column of arrows. The logistic and threshold models, constant- and effort-proportional harvesting, and the properties nonlinear equations lose all follow from this reading.\n",{"path":15486,"title":15487,"module":15471,"summary":15488},"\u002Fdifferential-equations\u002Ffirst-order\u002Fexistence-uniqueness-euler","Existence, Uniqueness, and Euler's Method","Existence and uniqueness can be settled before any attempt to solve. The existence-uniqueness theorem gives sufficient conditions on f, and a standard example shows what fails when they do not hold. Picard's successive approximations build the solution as the limit of an iteration, and Euler's method turns the same tangent-line idea into a numerical procedure for the equations no formula reaches.\n",{"path":15490,"title":15491,"module":15471,"summary":15492},"\u002Fdifferential-equations\u002Ffirst-order\u002Ffirst-order-difference-equations","First-Order Difference Equations","A difference equation advances a sequence one index at a time by a rule y_{n+1} = f(y_n). The linear case y_{n+1} = rho*y_n + b solves in closed form and converges to its equilibrium exactly when the ratio has magnitude below one, which underlies compound-interest and loan calculations. The logistic difference equation shows the nonlinear counterpart: an exchange of stability, a cascade of period doublings, and the onset of chaos.\n",{"path":15494,"title":15495,"module":15496,"summary":15497},"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fhomogeneous-constant-coefficients","Homogeneous Equations, the Wronskian, and Real Roots","Second-Order Linear Equations","A second-order linear homogeneous equation with constant coefficients is solved by guessing an exponential and reducing to the quadratic characteristic equation. Two solutions span every solution exactly when their Wronskian is nonzero; that condition, superposition, and Abel's formula give the full structure of the general solution for the case of two distinct real roots.\n",{"path":15499,"title":15500,"module":15496,"summary":15501},"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fcomplex-and-repeated-roots","Complex Roots, Repeated Roots, and Reduction of Order","When the characteristic equation has complex conjugate roots, Euler's formula converts the complex exponentials into a real fundamental set of decaying or growing oscillations. When it has a repeated root, one exponential is lost and reduction of order recovers the missing second solution as $t\\,e^{rt}$. The same substitution $y = v(t)y_1(t)$ finds a second solution from any known one.\n",{"path":15503,"title":15504,"module":15496,"summary":15505},"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fnonhomogeneous-undetermined-coefficients","Nonhomogeneous Equations: Undetermined Coefficients","The general solution of a nonhomogeneous linear equation is a complementary solution plus any one particular solution. When the forcing term is a polynomial, exponential, sine, or cosine, a particular solution can be found by assuming a trial form of the same shape with unknown coefficients and solving for them. The one complication is resonance, handled by multiplying the trial by a power of $t$.\n",{"path":15507,"title":15508,"module":15496,"summary":15509},"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fvariation-of-parameters","Variation of Parameters","Variation of parameters finds a particular solution of any nonhomogeneous linear equation from a fundamental set of the homogeneous one. Replacing the constants in the complementary solution by functions and imposing one convenient constraint reduces the problem to a two-by-two linear system whose solution is expressed through the Wronskian, giving an integral formula that works for forcing terms undetermined coefficients cannot touch.\n",{"path":15511,"title":15512,"module":15496,"summary":15513},"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fmechanical-electrical-vibrations","Mechanical and Electrical Vibrations","A spring-mass-damper obeys a second-order linear equation, and so does a series RLC circuit, with the same mathematics governing both. Free undamped motion is a pure sinusoid; damping adds a decaying envelope with three regimes; periodic forcing produces a transient that dies out and a steady-state oscillation whose amplitude peaks sharply near the natural frequency, the phenomenon of resonance.\n",{"path":15515,"title":15516,"module":15496,"summary":15517},"\u002Fdifferential-equations\u002Fsecond-order-linear\u002Fhigher-order-linear","Higher-Order Linear Equations","The second-order theory extends directly to order $n$: the solution space is $n$-dimensional, spanned by any $n$ solutions with nonzero Wronskian. For constant coefficients the characteristic polynomial has degree $n$, and its roots (counted with multiplicity, real and complex) build the basis by the same rules as before. Coupled oscillators are the natural application that raises the order.\n",{"path":15519,"title":15520,"module":15521,"summary":15522},"\u002Fdifferential-equations\u002Fseries-solutions\u002Fpower-series-ordinary-points","Power Series Solutions Near Ordinary Points","Series Solutions and Special Functions","A linear equation with variable coefficients has no characteristic equation. A power series substituted into the equation matches coefficients to a recurrence relation, which near an ordinary point yields two independent analytic solutions. The radius of convergence is at least the distance from the expansion point to the nearest singular point in the complex plane.\n",{"path":15524,"title":15525,"module":15521,"summary":15526},"\u002Fdifferential-equations\u002Fseries-solutions\u002Fregular-singular-frobenius","Euler Equations, Regular Singular Points, and Frobenius","The Euler equation x^2 y'' + a x y' + b y = 0 is solved outright by y = x^r, and its three root cases fix the behavior at any regular singular point. The Frobenius method multiplies x^r by a power series; the indicial equation chooses the exponents, and equal or integer-separated roots force a logarithm in the second solution. Gauss's hypergeometric equation is the archetype containing most classical functions as special cases.\n",{"path":15528,"title":15529,"module":15521,"summary":15530},"\u002Fdifferential-equations\u002Fseries-solutions\u002Fbessel-and-special-functions","Bessel's Equation, Legendre Polynomials, and Special Functions","Bessel's equation puts the Frobenius machinery through all three of its cases and produces the functions J and Y that govern anything vibrating or diffusing with circular symmetry. The gamma function extends the factorial so that Bessel functions of every order make sense; Legendre's equation, run through the hypergeometric form, yields the polynomials that play the same role in spherical geometry. Orthogonality ties both families to the eigenfunction expansions of Sturm–Liouville theory.\n",{"path":15532,"title":15533,"module":15534,"summary":15535},"\u002Fdifferential-equations\u002Flaplace\u002Flaplace-definition-ivps","The Laplace Transform: Definition, Properties, and Solving IVPs","The Laplace Transform","The Laplace transform sends a function of time to a function of a complex frequency by integrating it against the kernel e^{-st}. Differentiation in t becomes multiplication by s, so a linear constant-coefficient initial value problem turns into an algebraic equation. Existence rests on piecewise continuity and exponential order; the derivative rule folds in the initial data; and inversion runs through a transform table and partial fractions.\n",{"path":15537,"title":15538,"module":15534,"summary":15539},"\u002Fdifferential-equations\u002Flaplace\u002Fstep-impulse-convolution","Step Functions, Discontinuous Forcing, Impulses, and Convolution","The Heaviside step function and the second shifting theorem transform switches and discontinuous forcing into exponential factors on the transform. The Dirac delta idealizes an instantaneous impulse and transforms to a pure exponential. The convolution theorem inverts a product of transforms, writes the forced response as the impulse response convolved with the input, and solves Abel's tautochrone by transform.\n",{"path":15541,"title":15542,"module":15543,"summary":15544},"\u002Fdifferential-equations\u002Fsystems\u002Fmatrices-eigenvalues-review","Matrices, Linear Systems, and the Eigenvalue Toolkit","Systems of First-Order Linear Equations","Any nth-order linear equation, and any coupled collection of them, rewrites as a single first-order system x' = P(t)x + g(t). The matrix and vector algebra behind that form, the eigenvalue problem det(A - λI) = 0 that drives every solution method, and the fundamental theory — superposition, the Wronskian, Abel's theorem — together establish that n independent solutions span all solutions.\n",{"path":15546,"title":15547,"module":15543,"summary":15548},"\u002Fdifferential-equations\u002Fsystems\u002Fconstant-coefficient-systems-phase-portraits","Homogeneous Constant-Coefficient Systems and Phase Portraits","For x' = Ax with A constant, the trial x = ξe^{rt} turns the differential equation into the eigenvalue problem Aξ = rξ. The eigenvalues fix the geometry of the phase plane: real opposite signs give a saddle, real same sign a node, complex a spiral, purely imaginary a center. Worked in the plane, these cases form the eigenvalue-type classification of equilibria.\n",{"path":15550,"title":15551,"module":15543,"summary":15552},"\u002Fdifferential-equations\u002Fsystems\u002Frepeated-eigenvalues-fundamental-matrices","Repeated Eigenvalues, Fundamental Matrices, and Nonhomogeneous Systems","When a repeated eigenvalue supplies too few eigenvectors, a generalized eigenvector supplies the missing solution as ξte^{ρt} + ηe^{ρt}, giving an improper node. A fundamental set packaged as a matrix Φ(t) yields the matrix exponential e^{At}, the propagator mapping initial states to later ones. Variation of parameters solves the nonhomogeneous system x' = Ax + g(t).\n",{"path":15554,"title":15555,"module":15556,"summary":15557},"\u002Fdifferential-equations\u002Fnumerical\u002Feuler-and-runge-kutta","Euler, Improved Euler, and Runge–Kutta","Numerical Methods","Most initial value problems have no closed-form solution, so the solution is approximated on a grid. Euler's method steps along the tangent line, the improved Euler method averages two slopes, and the classical Runge–Kutta method averages four. Each added stage raises the order of accuracy at the cost of more evaluations per step, measured by how the local and global truncation errors scale with the step size.\n",{"path":15559,"title":15560,"module":15556,"summary":15561},"\u002Fdifferential-equations\u002Fnumerical\u002Fmultistep-systems-stability","Multistep Methods, Systems, and Stability","One-step methods discard everything but the last point. Multistep methods fit a polynomial to several past values and integrate it forward: the explicit Adams–Bashforth formulas, the implicit and more accurate Adams–Moulton formulas, and predictor–corrector pairs that combine them. The same rules extend verbatim to systems in vector form. A separate concern is stability: round-off can dominate truncation, and stiff equations force a tiny step for stability even when accuracy would allow a large one.\n",{"path":15563,"title":15564,"module":15565,"summary":15566},"\u002Fdifferential-equations\u002Fnonlinear\u002Fphase-plane-autonomous-stability","The Phase Plane, Critical Points, and Stability","Nonlinear Systems and Stability","Most nonlinear systems cannot be solved in closed form, so they are studied geometrically. The phase plane turns an autonomous planar system into a family of trajectories; the five archetypes of critical point follow from the eigenvalues of the coefficient matrix; the trace-determinant plane reads off type and stability directly; and epsilon-delta definitions make stability, asymptotic stability, and instability precise.\n",{"path":15568,"title":15569,"module":15565,"summary":15570},"\u002Fdifferential-equations\u002Fnonlinear\u002Flocally-linear-and-liapunov","Locally Linear Systems and Liapunov's Method","Near a critical point a nonlinear system looks linear, and the linear part is the Jacobian. The linearization fixes the type and stability of the nonlinear critical point in every case except a center or a repeated eigenvalue. Liapunov's direct method settles those cases and bounds the basin of attraction by constructing an energy-like function, without solving the system.\n",{"path":15572,"title":15573,"module":15565,"summary":15574},"\u002Fdifferential-equations\u002Fnonlinear\u002Fcompeting-species-predator-prey-limit-cycles","Population Models, Limit Cycles, and Chaos","The phase-plane methods apply directly to interacting-population models. Competing species either coexist or drive one another to extinction, decided by a single inequality among the interaction constants; the Lotka-Volterra predator-prey system produces closed population cycles. Limit cycles and the Poincaré-Bendixson theorem, the van der Pol oscillator, and the Lorenz equations with their strange attractor carry the theory into chaos.\n",{"path":15576,"title":15577,"module":15578,"summary":15579},"\u002Fdifferential-equations\u002Fpdes-fourier-bvp\u002Ffourier-series","Fourier Series and Convergence","PDEs, Fourier Series, and Boundary Value Problems","A two-point boundary value problem has nontrivial solutions only at a discrete set of eigenvalues, the same trichotomy that governs a singular linear system. For y'' + lambda y = 0 with zero endpoints the eigenfunctions are sines and cosines, and their orthogonality gives the Euler-Fourier coefficient formulas. The convergence theorem fixes when the series returns the function, the Gibbs phenomenon measures the overshoot at a jump, and even\u002Fodd symmetry produces half-range sine and cosine series.\n",{"path":15581,"title":15582,"module":15578,"summary":15583},"\u002Fdifferential-equations\u002Fpdes-fourier-bvp\u002Fheat-wave-laplace-equations","Separation of Variables: Heat, Wave, and Laplace Equations","Separation of variables replaces a partial differential equation by a pair of ordinary ones joined through a shared separation constant. Applied to the heat equation it produces the eigenvalue problem X'' + lambda X = 0, and the solution assembles as a Fourier series in the eigenfunctions. The same steps solve the wave equation, whose modes are standing waves, and Laplace's equation, the steady-state limit posed on a region rather than an interval.\n",{"path":15585,"title":15586,"module":15578,"summary":15587},"\u002Fdifferential-equations\u002Fpdes-fourier-bvp\u002Fsturm-liouville","Sturm-Liouville Theory","The eigenvalue problem behind separation of variables generalizes to the self-adjoint Sturm-Liouville form. Lagrange's identity makes the operator symmetric, and from that one fact follow real eigenvalues, orthogonal eigenfunctions, and eigenfunction expansions that behave like Fourier series. Singular problems admit Bessel and Legendre functions, and Sturm's separation and comparison theorems describe how the eigenfunctions oscillate.\n",{"path":15589,"title":15590,"module":15591,"summary":15592},"\u002Fdifferential-equations\u002Fhistory-variations\u002Fcalculus-of-variations","The Calculus of Variations","Historical Notes and the Calculus of Variations","Ordinary calculus finds the point where a function is stationary; the calculus of variations finds the whole curve where an integral is stationary. Euler's differential equation is the necessary condition for an extremal, and it becomes integrable in three cases, solving the shortest-path, minimal-surface, and brachistochrone problems. Lagrange multipliers extend the method to isoperimetric constraints, and Hamilton's principle recovers Newton's law from a single stationary integral.\n",{"path":15594,"title":15595,"module":15591,"summary":15596},"\u002Fdifferential-equations\u002Fhistory-variations\u002Fhistorical-notes","Great Problems and the People Who Solved Them","Differential equations grew out of specific problems, not a plan: the invention of calculus by Newton and Leibniz, the Bernoulli brachistochrone challenge, Euler's flood of methods, Lagrange's analytical mechanics, Gauss and Riemann's rigor, Laplace's celestial mechanics, and Poincaré's qualitative theory. Each method descends from a named problem, and reading the subject forward from those problems explains why its parts fit together.\n",{"path":15598,"title":15599,"module":6,"summary":6},"\u002Fdifferential-equations","Differential Equations",{"path":15601,"title":15602,"module":15603,"summary":15604},"\u002Frelativity\u002Ffoundations\u002Fspecial-relativity-postulates","The Postulates of Special Relativity","Foundations of Relativity","Newton's laws are the same in every inertial frame, but Maxwell's are not: the equations of electromagnetism single out one speed, c, and the nineteenth century read that as the speed of light relative to a medium, the ether. The Michelson-Morley experiment looked for Earth's motion through that medium and found nothing. Einstein's two postulates replace the ether, and their first consequence is that simultaneity is frame-dependent.\n",{"path":15606,"title":15607,"module":15603,"summary":15608},"\u002Frelativity\u002Ffoundations\u002Florentz-transformation-spacetime","The Lorentz Transformation and Spacetime","Requiring that a light sphere stay a light sphere in every inertial frame fixes the coordinate change between frames uniquely: the Lorentz transformation, with its factor gamma. Differentiating it gives relativistic velocity addition, which caps composed speeds at c. Plotting the same events on skewed spacetime axes turns the algebra into geometry, with calibration hyperbolae, an invariant interval, and a light cone that sorts events into past, future, and elsewhere.\n",{"path":15610,"title":15611,"module":15603,"summary":15612},"\u002Frelativity\u002Ffoundations\u002Ftime-dilation-length-contraction","Time Dilation, Length Contraction, and Paradoxes","A light clock and the constancy of c give the two headline effects directly: a moving clock runs slow by gamma, and a moving rod is short by the same factor. Cosmic-ray muons reaching sea level are the standing experimental proof. The relativistic Doppler effect adds the time-dilation factor to the classical shift, and the twin and pole-barn paradoxes dissolve once the relativity of simultaneity is taken seriously.\n",{"path":15614,"title":15615,"module":15603,"summary":15616},"\u002Frelativity\u002Ffoundations\u002Frelativistic-momentum-energy","Relativistic Momentum and Energy","Conserving momentum in every inertial frame forces the redefinition p = gamma m u, which diverges as the speed approaches c. Integrating the corresponding force gives the total energy E = gamma m c-squared, whose rest term m c-squared is Einstein's mass-energy equivalence. Energy and momentum join into a four-vector whose invariant length is the rest energy, giving E-squared = (pc)-squared + (m c-squared)-squared, massless particles, and nuclear binding energy.\n",{"path":15618,"title":15619,"module":15603,"summary":15620},"\u002Frelativity\u002Ffoundations\u002Fgeneral-relativity","A Taste of General Relativity","Einstein's happiest thought was that a freely falling observer feels no gravity: a uniform gravitational field is locally indistinguishable from an accelerating frame. That equivalence principle predicts that light bends near a mass, that clocks run slow deep in a gravitational well, that Mercury's orbit precesses, and that radar echoes are delayed. Every prediction has been confirmed, and pushing the redshift to its limit gives the black hole.\n",{"path":15622,"title":15623,"module":15624,"summary":15625},"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Fminkowski-spacetime-and-the-interval","Minkowski Spacetime and the Interval","Spacetime and the Lorentz Group","The Lorentz transformation of the foundations module is repackaged as the geometry of a four-dimensional space whose invariant is not a distance but the spacetime interval. Events, worldlines, and the metric signature define a causal structure that every observer shares. Proper time is the length of a timelike worldline, and the twin paradox becomes the statement that a straight worldline accumulates the most proper time.\n",{"path":15627,"title":15628,"module":15624,"summary":15629},"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Ffour-vectors-and-index-notation","Four-Vectors and Index Notation","The index calculus that the rest of the course runs on. Contravariant and covariant components, the Minkowski metric as the machine that raises and lowers indices, and the Einstein summation convention are assembled into scalar products that are the same in every frame. The four-velocity and four-acceleration follow, together with the identity that the four-velocity has constant invariant length.\n",{"path":15631,"title":15632,"module":15624,"summary":15633},"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Fthe-lorentz-group-and-rapidity","The Lorentz Group and Rapidity","The Lorentz transformations are the linear maps that preserve the Minkowski metric, and they form the group O(1,3). Boosts are hyperbolic rotations parametrized by rapidity, which adds along a line where velocity does not. The boost and rotation generators fix the group's local structure; its four disconnected components are set by two signs; and two non-collinear boosts compose into a boost plus a rotation, the Wigner rotation behind Thomas precession.\n",{"path":15635,"title":15636,"module":15624,"summary":15637},"\u002Frelativity\u002Fspacetime-and-the-lorentz-group\u002Fdoppler-aberration-and-appearance","Doppler, Aberration, and Appearance","Light carries a null four-momentum, and boosting it produces every optical effect of relativity at once. The covariant Doppler formula follows from the transformation of frequency, aberration from the transformation of direction, and the headlight effect from the resulting concentration of light forward. The Terrell-Penrose result shows that a fast object photographs as rotated, not contracted.\n",{"path":15639,"title":15640,"module":15641,"summary":15642},"\u002Frelativity\u002Frelativistic-dynamics\u002Ffour-momentum-force-and-accelerated-motion","Four-Momentum, Four-Force, and Accelerated Motion","Relativistic Dynamics","The four-momentum packages energy and momentum into a single vector whose invariant length is the rest mass. Its proper-time derivative is the four-force, always orthogonal to the four-velocity, and a constant orthogonal four-force produces hyperbolic motion. Constant proper acceleration gives rapidity linear in proper time, the relativistic rocket equation, and the Rindler horizon behind an eternally accelerating observer.\n",{"path":15644,"title":15645,"module":15641,"summary":15646},"\u002Frelativity\u002Frelativistic-dynamics\u002Fparticle-decays-and-two-body-kinematics","Particle Decays and Two-Body Kinematics","Conservation of four-momentum fixes the kinematics of a decay from the masses alone. In the center-of-momentum frame a parent breaks into two daughters with equal and opposite momenta and energies set by the Kallen triangle function. Boosting to the lab opens the decay into a cone, and the invariant mass built from the daughters reconstructs the parent as a peak. Worked cases: the two-photon decay of the neutral pion and a heavy two-body hadronic decay.\n",{"path":15648,"title":15649,"module":15641,"summary":15650},"\u002Frelativity\u002Frelativistic-dynamics\u002Fcollisions-thresholds-and-the-cm-frame","Relativistic Collisions and Threshold Energies","Two-body collisions run on the same conserved four-momentum as decays. The invariant s sets the total energy available in the center-of-momentum frame and therefore the threshold for producing new particles. Fixed-target energy grows only as the square root of beam energy while a collider grows linearly, which is why colliders reach high energy. Compton scattering follows as a worked photon-electron collision giving the wavelength shift.\n",{"path":15652,"title":15653,"module":15641,"summary":15654},"\u002Frelativity\u002Frelativistic-dynamics\u002Fmandelstam-variables-and-invariants","Mandelstam Variables and Lorentz Invariants","For a two-to-two process the three Mandelstam invariants s, t, and u encode all the kinematics in frame-independent form. They obey a single linear constraint, the sum of the four squared masses, so only two are independent. s is the center-of-momentum energy squared, t and u are momentum transfers tied to the scattering angle, and crossing symmetry relates one amplitude across three channels through these variables.\n",{"path":15656,"title":15657,"module":15658,"summary":15659},"\u002Frelativity\u002Fcovariant-electrodynamics\u002Ffour-current-and-the-four-potential","The Four-Current and Four-Potential","Covariant Electromagnetism","Charge density and current combine into a single four-vector whose divergence is charge conservation. The scalar and vector potentials combine likewise into the four-potential, whose gauge freedom fixes to the Lorenz condition, reducing Maxwell's equations for the potentials to a single wave equation sourced by the four-current.\n",{"path":15661,"title":15662,"module":15658,"summary":15663},"\u002Frelativity\u002Fcovariant-electrodynamics\u002Fthe-electromagnetic-field-tensor","The Electromagnetic Field Tensor","The antisymmetric derivative of the four-potential is the field-strength tensor F, gauge invariant by construction, with the electric and magnetic fields as its components. Its dual exchanges E and B, and its two contractions form the Lorentz invariants that classify a field as electric, magnetic, or radiative in every frame.\n",{"path":15665,"title":15666,"module":15658,"summary":15667},"\u002Frelativity\u002Fcovariant-electrodynamics\u002Ftransformation-of-electric-and-magnetic-fields","How E and B Transform","Transforming the field tensor under a boost gives explicit rules for the electric and magnetic fields: components along the motion are unchanged, transverse components mix and pick up a gamma. The field of a uniformly moving charge compresses transversely, and the force between a current and a moving charge shows that magnetism is the relativistic shadow of electrostatics.\n",{"path":15669,"title":15670,"module":15658,"summary":15671},"\u002Frelativity\u002Fcovariant-electrodynamics\u002Fcovariant-maxwell-and-the-stress-energy-tensor","Covariant Maxwell and the Stress–Energy Tensor","Maxwell's four equations collapse into two tensor equations, one sourced by the four-current and one an identity on the field strength, with charge conservation automatic. The Lorentz force becomes a four-vector law, and the field's energy, momentum, and stress assemble into a symmetric, conserved stress–energy tensor — the object that will source gravity.\n",{"path":15673,"title":15674,"module":15675,"summary":15676},"\u002Frelativity\u002Fcurved-spacetime\u002Fthe-equivalence-principle-formalized","The Equivalence Principle","Curved Spacetime","The equality of gravitational and inertial mass promotes to a physical principle in three graded strengths — weak, Einstein, and strong. A freely falling laboratory is locally indistinguishable from an inertial frame, but the qualifier \"locally\" is essential: the size of the patch over which gravity vanishes is set by the tidal field, which no change of frame can remove. Tidal forces are the true, coordinate-independent signature of gravity, and they are what curvature will measure.\n",{"path":15678,"title":15679,"module":15675,"summary":15680},"\u002Frelativity\u002Fcurved-spacetime\u002Fmanifolds-vectors-and-the-metric","Manifolds, Vectors, and the Metric","A manifold is a space that looks locally like flat space, described by overlapping coordinate charts. Tangent vectors are directional derivatives with the coordinate basis vectors as partial-derivative operators; one-forms live in the dual space; and the metric tensor turns a coordinate line element into an invariant length. The 2-sphere and Rindler metrics serve as worked examples, including the coordinate singularities that are artefacts of the chart, not of the geometry.\n",{"path":15682,"title":15683,"module":15675,"summary":15684},"\u002Frelativity\u002Fcurved-spacetime\u002Fcovariant-derivative-and-christoffel-symbols","Parallel Transport and the Covariant Derivative","The ordinary derivative of a vector field is not a tensor, because it subtracts vectors living in different tangent spaces. A connection supplies the missing comparison: the covariant derivative adds Christoffel-symbol correction terms that cancel the coordinate artefacts. Requiring the connection to be torsion-free and to preserve the metric fixes the Christoffel symbols uniquely in terms of derivatives of the metric, giving the Levi-Civita connection that general relativity uses.\n",{"path":15686,"title":15687,"module":15675,"summary":15688},"\u002Frelativity\u002Fcurved-spacetime\u002Fgeodesics-and-the-geodesic-equation","Geodesics and the Newtonian Limit","Free fall is geodesic motion: a freely falling particle follows the straightest possible worldline, obtained either by parallel-transporting its own tangent vector or by extremizing proper time. Both routes give the geodesic equation. Affine parameters, and conserved quantities from symmetries via Killing vectors, make it solvable. In the weak-field slow-motion limit the geodesic equation reproduces Newton's law of gravity, fixing the time-time metric component as the Newtonian potential.\n",{"path":15690,"title":15691,"module":15675,"summary":15692},"\u002Frelativity\u002Fcurved-spacetime\u002Fcurvature-riemann-and-geodesic-deviation","Curvature and the Riemann Tensor","Curvature is the failure of parallel transport to commute: carrying a vector around an infinitesimal loop returns it rotated, and the rotation per unit area is the Riemann tensor. Its symmetries cut the components to twenty in four dimensions. Geodesic deviation makes it the equation of tidal forces, and its contractions — the Ricci tensor, the Ricci scalar, and the divergence-free Einstein tensor — assemble the objects the field equation is built from.\n",{"path":15694,"title":15695,"module":15675,"summary":15696},"\u002Frelativity\u002Fcurved-spacetime\u002Fthe-einstein-field-equations","The Einstein Field Equations","The field equation is assembled from a short list of requirements: a symmetric, divergence-free, second-order geometric tensor set proportional to the stress–energy tensor, with the coefficient fixed by the Newtonian limit. The cosmological constant is the one extra term the requirements allow. The Einstein–Hilbert action gives the same equation from a variational principle, and the coupled system closes the logic of the module: matter curves spacetime, and spacetime tells matter how to move.\n",{"path":15698,"title":15699,"module":15700,"summary":15701},"\u002Frelativity\u002Fthe-schwarzschild-solution\u002Fthe-schwarzschild-metric","The Schwarzschild Metric","The Schwarzschild Solution","The first exact solution of Einstein's equation follows from two assumptions, staticity and spherical symmetry, imposed on the vacuum outside a mass. Solving the vacuum field equations fixes two metric functions and produces the Schwarzschild geometry, whose one length scale is the Schwarzschild radius $r_s = 2GM\u002Fc^2$. Birkhoff's theorem shows this is the only spherical vacuum, and the far field reduces to Newtonian gravity.\n",{"path":15703,"title":15704,"module":15700,"summary":15705},"\u002Frelativity\u002Fthe-schwarzschild-solution\u002Fgeodesics-and-orbits-in-schwarzschild","Orbits in the Schwarzschild Geometry","The two Killing symmetries of the Schwarzschild metric give a conserved energy and angular momentum per unit mass, reducing geodesic motion to a one-dimensional problem in an effective potential. The potential carries an extra attractive $1\u002Fr^3$ term absent from Newton's, which caps the centrifugal barrier, produces an innermost stable circular orbit at $6GM\u002Fc^2$, and makes bound orbits precess instead of closing.\n",{"path":15707,"title":15708,"module":15700,"summary":15709},"\u002Frelativity\u002Fthe-schwarzschild-solution\u002Flight-bending-and-null-geodesics","Null Geodesics and the Photon Sphere","Light follows null geodesics, governed by a photon effective potential with a single unstable maximum at $3GM\u002Fc^2$, the photon sphere. The impact parameter sorts rays into those that escape with a deflection and those captured, with the critical value $b_c = 3\\sqrt{3}\\,GM\u002Fc^2$ dividing them. A grazing ray bends by $4GM\u002F(c^2 b)$, twice the naive Newtonian value, and the critical impact parameter sets the edge of a black hole's shadow.\n",{"path":15711,"title":15712,"module":15713,"summary":15714},"\u002Frelativity\u002Ftests-of-general-relativity\u002Fperihelion-precession-of-mercury","The Perihelion Precession of Mercury","Tests of General Relativity","A single extra term in the Schwarzschild orbit equation, cubic in the inverse radius, keeps a bound orbit from closing. The perturbation advances the perihelion by 6πGM\u002F(c²a(1−e²)) per revolution, which for Mercury is 43 arcseconds per century — exactly the anomaly left after Newtonian planetary perturbations are subtracted. A note on frame dragging closes the lesson.\n",{"path":15716,"title":15717,"module":15713,"summary":15718},"\u002Frelativity\u002Ftests-of-general-relativity\u002Fdeflection-of-light-and-gravitational-lensing","Light Deflection and Gravitational Lensing","A light ray grazing the Sun bends by 4GM\u002F(c²b), exactly twice the value a Newtonian corpuscle would give; the extra factor is the curvature of space. The 1919 eclipse confirmed it. The same bending focuses light from distant sources into Einstein rings, multiple images, and microlensing brightenings, making lensing a direct probe of mass, including mass that emits no light.\n",{"path":15720,"title":15721,"module":15713,"summary":15722},"\u002Frelativity\u002Ftests-of-general-relativity\u002Fgravitational-redshift-and-shapiro-delay","Gravitational Redshift and the Shapiro Delay","A clock deeper in a gravitational well ticks slower, and a photon climbing out loses frequency by the ratio of the metric's time-time components. Pound and Rebka measured the 2.5×10⁻¹⁵ shift over a 22.5-metre tower. Radar signals grazing the Sun return late by about 250 microseconds, the Shapiro delay. Both probe the time part of the metric directly.\n",{"path":15724,"title":15725,"module":15713,"summary":15726},"\u002Frelativity\u002Ftests-of-general-relativity\u002Frelativity-in-technology-gps","Relativity and the Global Positioning System","A GPS satellite clock runs slow by 7 microseconds a day from its orbital speed and fast by 46 from its higher gravitational potential, a net gain of about 38 microseconds a day. Left uncorrected, the timing error would grow into kilometres of position error within a day and exceed navigation tolerance within minutes. The satellites carry a pre-launch frequency offset to cancel it.\n",{"path":15728,"title":15729,"module":15730,"summary":15731},"\u002Frelativity\u002Fblack-holes\u002Fhorizons-and-coordinate-singularities","Horizons and Coordinate Singularities","Black Holes","The Schwarzschild radius is a coordinate singularity, not a curvature singularity: the metric blows up there only because the static coordinates fail, while the geometry stays finite. Eddington–Finkelstein and Kruskal– Szekeres coordinates cross the horizon smoothly and show the light cones tipping toward the center. A freely falling observer reaches the true singularity at r=0 in finite proper time, while a distant observer sees the infall freeze and redden at the horizon.\n",{"path":15733,"title":15734,"module":15730,"summary":15735},"\u002Frelativity\u002Fblack-holes\u002Frotating-and-charged-black-holes","Rotating and Charged Black Holes","A stationary black hole is fixed by three numbers: mass, angular momentum, and charge. The Reissner–Nordström metric adds charge and splits the horizon in two; the Kerr metric adds rotation, drags inertial frames, and wraps the horizon in an ergosphere where nothing can stay still. Inside the ergosphere the Penrose process extracts rotational energy, and the no-hair theorem states that no other detail of the collapsed matter survives.\n",{"path":15737,"title":15738,"module":15730,"summary":15739},"\u002Frelativity\u002Fblack-holes\u002Fblack-hole-thermodynamics","Black-Hole Thermodynamics","The four laws of black-hole mechanics mirror the four laws of thermodynamics term for term, with horizon area playing the role of entropy and surface gravity the role of temperature. Hawking's calculation makes the analogy literal: a black hole radiates at a temperature set by its surface gravity, carries a real entropy proportional to its horizon area, and slowly evaporates. The thermal spectrum raises the information paradox.\n",{"path":15741,"title":15742,"module":15743,"summary":15744},"\u002Frelativity\u002Fgravitational-waves\u002Flinearized-gravity-and-wave-solutions","Linearized Gravity and Wave Solutions","Gravitational Waves","Weak gravity is a small perturbation of flat spacetime, and the linearized Einstein equation in the Lorenz gauge is an ordinary wave equation propagating at the speed of light. The trace-reversed perturbation carries the dynamics, residual gauge freedom fixes the transverse-traceless form, and the two physical polarizations deform a ring of freely falling masses into oscillating ellipses whose fractional size change is the strain.\n",{"path":15746,"title":15747,"module":15743,"summary":15748},"\u002Frelativity\u002Fgravitational-waves\u002Fgeneration-and-the-quadrupole-formula","The Quadrupole Formula","The retarded solution of the linearized field equation gives the field of a moving source, and conservation of mass and momentum forbids monopole and dipole radiation, leaving the mass quadrupole as the leading emitter. The quadrupole formula fixes the strain and the radiated luminosity, and applied to a compact binary it predicts the inspiral chirp of rising frequency and amplitude. The Hulse-Taylor pulsar's orbital decay confirmed it to a fraction of a percent.\n",{"path":15750,"title":15751,"module":15743,"summary":15752},"\u002Frelativity\u002Fgravitational-waves\u002Fdetection-ligo-and-the-first-events","LIGO and the First Detections","A gravitational wave is measured as a differential length change of the two arms of a kilometre-scale Michelson interferometer, a strain of order ten to the minus twenty-one that moves the mirrors by a fraction of a proton radius. GW150914 recorded the inspiral, merger, and ringdown of two black holes, fixing their masses and the energy radiated, and GW170817 with its coincident gamma-ray burst and kilonova opened multimessenger astronomy.\n",{"path":15754,"title":15755,"module":15756,"summary":15757},"\u002Frelativity\u002Fcosmological-bridge\u002Fthe-cosmological-principle-and-flrw-metric","The Cosmological Principle and the FLRW Metric","A Bridge to Cosmology","Homogeneity and isotropy restrict the spacetime of the universe to a single family of metrics: a flat cosmic-time slicing of spatial sections of constant curvature, scaled by a time-dependent factor a(t). This lesson builds the Friedmann–Lemaître–Robertson–Walker metric from those symmetries, separates comoving from proper distance, and derives cosmological redshift as the stretching of wavelengths with the scale factor.\n",{"path":15759,"title":15760,"module":15756,"summary":15761},"\u002Frelativity\u002Fcosmological-bridge\u002Ffriedmann-equations-and-cosmic-dynamics","The Friedmann Equations and Cosmic Dynamics","The Einstein equation applied to the FLRW metric with a perfect-fluid source yields the two Friedmann equations and the conservation law that ties them together. This lesson derives them, defines the critical density and the density parameters that fix the spatial geometry, works out how matter, radiation, and a cosmological constant dilute and drive the expansion, and hands off to a dedicated cosmology subject.\n",{"path":15763,"title":15764,"module":6,"summary":6},"\u002Frelativity","Relativity",{"path":15766,"title":15767,"module":6,"summary":6},"\u002Fphysical-computing","Physical Computing",{"path":15769,"title":15770,"module":15771,"summary":15772},"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fblackbody-radiation-and-the-planck-quantum","Blackbody Radiation and the Planck Quantum","Origins of the Quantum","Millikan's oil-drop experiment fixed the electron charge as an indivisible unit, and the spectrum of thermal radiation forced a second, deeper quantum. Classical physics predicts an infinite energy density at short wavelengths; Planck removed the divergence by allowing a cavity oscillator to hold only energies that are integer multiples of hf, the first appearance of the quantum of action.\n",{"path":15774,"title":15775,"module":15771,"summary":15776},"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fthe-photoelectric-effect-and-the-photon","The Photoelectric Effect and the Photon","Light shone on a clean metal ejects electrons, but the details defied the wave theory: the electrons' maximum energy depends on the light's frequency, not its brightness, and there is a sharp threshold frequency below which nothing happens. Einstein resolved every anomaly by treating light as a stream of energy quanta hf, each absorbed whole by one electron, and Millikan's measurement of the stopping-potential slope confirmed h to a decade before anyone expected.\n",{"path":15778,"title":15779,"module":15771,"summary":15780},"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fx-rays-and-the-compton-effect","X-Rays and the Compton Effect","X-rays are short-wavelength electromagnetic waves produced when fast electrons are braked in a target, and their diffraction by crystals lets Bragg's law measure atomic spacings. Compton then scattered X-rays off electrons and found the wavelength shifted by an amount that only a photon carrying momentum hf\u002Fc could explain, closing the case for the particle nature of light.\n",{"path":15782,"title":15783,"module":15771,"summary":15784},"\u002Fquantum-mechanics\u002Fold-quantum-theory\u002Fthe-old-quantum-theory-bohr-and-sommerfeld","The Old Quantum Theory: Bohr, Sommerfeld, and Correspondence","Between Bohr's 1913 atom and Schrödinger's 1926 equation, physics ran on a provisional recipe: keep classical orbits, but admit only those whose action integral is a whole multiple of Planck's constant. This lesson develops the Wilson-Sommerfeld phase-integral rule, applies it to the oscillator and to the elliptical Kepler orbits of hydrogen, derives Sommerfeld's relativistic fine structure and the quantization of orbit orientation, and shows how the correspondence principle fixed intensities and selection rules. The systematic failures — helium, line intensities, the anomalous Zeeman effect — mark exactly where a theory of orbits had to give way to a theory of waves.\n",{"path":15786,"title":15787,"module":15788,"summary":15789},"\u002Fquantum-mechanics\u002Fmatter-waves\u002Fde-broglie-waves-and-electron-diffraction","De Broglie Waves and Electron Diffraction","The Wave Nature of Matter","In 1924 de Broglie proposed that every particle carries a wave of wavelength h\u002Fp. The hypothesis explains Bohr's quantized orbits as standing waves, and Davisson and Germer, then G. P. Thomson, confirmed it by diffracting electrons from crystals exactly as X-rays diffract. We derive the electron wavelength, work the Bragg analysis of the data, and give the relativistic form.\n",{"path":15791,"title":15792,"module":15788,"summary":15793},"\u002Fquantum-mechanics\u002Fmatter-waves\u002Fwave-packets-and-the-probability-interpretation","Wave Packets and the Probabilistic Wave Function","A single de Broglie wave fills all space, but a particle is localized. Adding many waves of nearby wavelength builds a wave packet that is confined and moves at the group velocity, which equals the particle velocity. Born's rule reads the squared amplitude of the wave function as a probability density, the meaning confirmed by electron interference building up one detection at a time.\n",{"path":15795,"title":15796,"module":15788,"summary":15797},"\u002Fquantum-mechanics\u002Fmatter-waves\u002Fthe-uncertainty-principle","The Uncertainty Principle and Wave-Particle Duality","The packet relations delta-k delta-x about 1 become Heisenberg's principle once momentum is hbar times wave number: position and momentum cannot both be sharp, nor energy and time. The gamma-ray microscope shows the limit is physical, not technical. It fixes the zero-point energy of a confined particle, the size of the hydrogen atom, and the natural width of spectral lines, and it frames the wave-particle duality of all matter and radiation.\n",{"path":15799,"title":15800,"module":15801,"summary":15802},"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fthe-schrodinger-equation-in-one-dimension","The Schrödinger Equation in One Dimension","Wave Mechanics in One Dimension","The wave equation for matter cannot be derived; it is postulated and judged by experiment. We build the time-dependent Schrödinger equation from the de Broglie relations, read Born's probability rule off the complex wave function, and separate the time and space dependence to get the time-independent equation whose bound-state solutions are the stationary states. The five acceptability conditions on the wave function are what force energy to be quantized.\n",{"path":15804,"title":15805,"module":15801,"summary":15806},"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fthe-free-particle-and-wave-packet-dynamics","The Free Particle and Wave-Packet Dynamics","The free particle has no bound states: its stationary solutions are non-normalizable plane waves forming a continuum. Physical states are wave packets built by superposing them, and the superposition is a Fourier transform. We delta-normalize the plane waves, assemble a Gaussian packet, solve for its exact time evolution, and read off the two facts that reconcile the wave picture with mechanics: the packet moves at the group velocity ħk\u002Fm, the classical velocity, and it spreads because its component momenta travel at different speeds.\n",{"path":15808,"title":15809,"module":15801,"summary":15810},"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fparticle-in-infinite-and-finite-square-wells","Particle in Infinite and Finite Square Wells","The infinite square well is the simplest bound-state problem: two boundary conditions quantize the energy into a ladder E_n = n² E_1, and the eigenfunctions are the standing waves of a string fixed at both ends. Relaxing the walls to a finite depth lets the wave function leak into the classically forbidden region, keeps the number of bound states finite, and turns the eigenvalue condition into a transcendental equation solved graphically.\n",{"path":15812,"title":15813,"module":15801,"summary":15814},"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Foperators-expectation-values-and-the-harmonic-oscillator","Operators, Expectation Values, and the Harmonic Oscillator","Measurable quantities are extracted from the wave function as expectation values, and each observable is represented by an operator that acts between Ψ* and Ψ — position by multiplication, momentum by a derivative, energy by the Hamiltonian. Applied to the harmonic oscillator, the machinery yields evenly spaced levels E_n = (n+½)ℏω, Gaussian- times-Hermite eigenfunctions of definite parity, and the selection rule Δn = ±1.\n",{"path":15816,"title":15817,"module":15801,"summary":15818},"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fthe-dirac-delta-potential","The Dirac-Delta Potential: A Single Bound State and Scattering","A potential concentrated at a single point is solvable in closed form and isolates the physics of matching a wave function across a discontinuity. Integrating the Schrödinger equation across the spike gives a jump condition on the derivative; the attractive delta well then supports exactly one bound state, of energy set by the strength alone, while the same spike scatters an incoming beam with a transmission that rises from zero to one. The attractive well and the repulsive barrier scatter identically yet only the well binds.\n",{"path":15820,"title":15821,"module":15801,"summary":15822},"\u002Fquantum-mechanics\u002Fwave-mechanics-1d\u002Fbarrier-penetration-and-quantum-tunneling","Barrier Penetration and Quantum Tunneling","Unbound states scatter rather than bind. A particle meeting a step is partly reflected even when it has more than enough energy to pass, and a particle meeting a barrier taller than its energy has a nonzero chance of appearing on the far side. Matching the wave function across the boundaries gives the reflection and transmission coefficients and the exponential tunneling probability that explains alpha decay, the scanning tunneling microscope, and the ammonia clock.\n",{"path":15824,"title":15825,"module":15826,"summary":15827},"\u002Fquantum-mechanics\u002Fformalism\u002Fhilbert-space-and-dirac-notation","Hilbert Space and Dirac Bra–Ket Notation","The Formalism of Quantum Mechanics","Wave mechanics is one representation of a deeper structure: quantum states are vectors in a complex inner-product space, and observables act on them as linear operators. We build that space from the axioms, introduce Dirac's kets and bras as vectors and the linear functionals that measure them, and identify the wavefunction as the components of an abstract state in the position basis. The resolution of the identity is the single algebraic tool that ties every basis, expansion, and matrix element together.\n",{"path":15829,"title":15830,"module":15826,"summary":15831},"\u002Fquantum-mechanics\u002Fformalism\u002Fobservables-hermitian-operators-and-eigenvalues","Observables, Hermitian Operators, and the Spectral Theorem","Every measurable quantity is represented by a Hermitian operator, and the reason is forced: a measurement needs real eigenvalues, orthogonal eigenvectors, and a complete eigenbasis, and Hermiticity delivers precisely those. We derive those properties from self-adjointness, state the spectral theorem, handle degeneracy, and show that two observables share an eigenbasis precisely when they commute — the algebraic condition behind compatible and incompatible measurements.\n",{"path":15833,"title":15834,"module":15826,"summary":15835},"\u002Fquantum-mechanics\u002Fformalism\u002Fthe-postulates-and-quantum-measurement","The Postulates and Quantum Measurement","With states as vectors and observables as Hermitian operators, the physical content of quantum mechanics reduces to a short list of postulates. We state them precisely, derive the Born probability rule for discrete and continuous spectra, work out projective collapse and its idempotence, compute expectation values and their variance, and state the measurement problem cleanly — the one place the postulates split unitary evolution from measurement without explaining the seam.\n",{"path":15837,"title":15838,"module":15826,"summary":15839},"\u002Fquantum-mechanics\u002Fformalism\u002Fposition-momentum-and-continuous-spectra","Position, Momentum, and Continuous Spectra","Position and momentum are the observables with no normalizable eigenstates: their spectra are continuous, their eigenkets are delta-normalized, and the two are Fourier conjugates. We derive the canonical commutator from the momentum operator, build the continuous-basis machinery (Dirac deltas replacing Kronecker deltas), show the position and momentum wavefunctions are a Fourier-transform pair, and compute expectation values in either representation.\n",{"path":15841,"title":15842,"module":15826,"summary":15843},"\u002Fquantum-mechanics\u002Fformalism\u002Fcommutators-and-the-generalized-uncertainty-principle","Commutators and the Generalized Uncertainty Principle","The commutator of two observables measures the obstruction to sharing an eigenbasis, and it bounds how sharply both can be known at once. We derive the generalized uncertainty relation from the Schwarz inequality, recover the position–momentum bound as a special case, characterize the minimum-uncertainty states that saturate it as Gaussians, and give the energy–time relation its correct reading as a lifetime bound rather than a commutator relation.\n",{"path":15845,"title":15846,"module":15826,"summary":15847},"\u002Fquantum-mechanics\u002Fformalism\u002Ftime-evolution-schrodinger-and-heisenberg-pictures","Time Evolution, Propagators, and the Heisenberg Picture","Time evolution is generated by the Hamiltonian and implemented by a unitary operator that preserves probability. We build that operator, expand a state in stationary states to see why probability densities freeze while phases wind, introduce the propagator, transfer the time dependence onto operators in the Heisenberg picture, and derive Ehrenfest's theorem — which recovers classical equations of motion for expectation values and identifies conserved quantities as observables commuting with the Hamiltonian.\n",{"path":15849,"title":15850,"module":15851,"summary":15852},"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fladder-operators-and-the-number-states","Ladder Operators and the Number States","The Oscillator Algebraically, and Symmetry","The harmonic oscillator can be solved without touching a differential equation. Factoring the Hamiltonian into a lowering operator and its adjoint turns the spectrum into pure algebra: the commutator relation fixes the ladder, the vacuum condition fixes the ground state, and the energies fall out as equally spaced rungs. The same operators give the matrix elements of position and momentum for free.\n",{"path":15854,"title":15855,"module":15851,"summary":15856},"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fcoherent-and-squeezed-states","Coherent and Squeezed States","A single number state never moves — its position expectation is pinned at the origin. The superposition that oscillates like a classical particle is the eigenstate of the annihilation operator: the coherent state. It is a displaced vacuum, carries Poissonian photon statistics, saturates the uncertainty bound, and traces a rigid Gaussian orbit in phase space. Squeezing deforms that circle, trading precision in one quadrature for noise in the other.\n",{"path":15858,"title":15859,"module":15851,"summary":15860},"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fsymmetries-generators-and-conservation-laws","Symmetries, Generators, and Conservation Laws","Every continuous symmetry of a quantum system is a unitary operator built by exponentiating a Hermitian generator: momentum generates translations, angular momentum generates rotations, the Hamiltonian generates time evolution. When a generator commutes with the Hamiltonian, the transformation leaves the dynamics unchanged and the generator is conserved — the quantum form of Noether's theorem — and any symmetry that mixes states within a level forces degeneracy.\n",{"path":15862,"title":15863,"module":15851,"summary":15864},"\u002Fquantum-mechanics\u002Foscillator-and-symmetry\u002Fparity-time-reversal-and-discrete-symmetries","Parity, Time Reversal, and Discrete Symmetries","Parity and time reversal are symmetries no continuous generator can reach. Parity is a unitary involution whose eigenvalues label states even or odd, fixing the dipole selection rules. Time reversal is antiunitary: it conjugates i, flips momenta and spins, and for half-integer spin squares to minus one, which by Kramers' theorem makes every level of a time-reversal-invariant Hamiltonian at least doubly degenerate.\n",{"path":15866,"title":15867,"module":14706,"summary":15868},"\u002Fquantum-mechanics\u002Fangular-momentum\u002Forbital-angular-momentum-and-spherical-harmonics","Orbital Angular Momentum and Spherical Harmonics","Orbital angular momentum is the operator triple built from position and momentum. Its components fail to commute, so no state carries sharp values of more than one of them, but each commutes with the total square. Solving the common eigenvalue problem in spherical coordinates quantizes both the magnitude and the projection and produces the spherical harmonics, the angular part of every central-force wavefunction.\n",{"path":15870,"title":15871,"module":14706,"summary":15872},"\u002Fquantum-mechanics\u002Fangular-momentum\u002Fthe-angular-momentum-algebra","The Angular-Momentum Algebra and Ladder Operators","The eigenvalues of angular momentum follow from the commutation relations alone, with no reference to coordinates or wavefunctions. Raising and lowering operators built from the components generate finite multiplets, force the total quantum number to be a non-negative integer or half-integer, and fix the matrix elements of every component. The half-integer values excluded by orbital motion appear here, and they are what spin realizes.\n",{"path":15874,"title":15875,"module":14706,"summary":15876},"\u002Fquantum-mechanics\u002Fangular-momentum\u002Faddition-of-angular-momenta-and-clebsch-gordan","Addition of Angular Momenta and Clebsch–Gordan Coefficients","Two angular momenta combine into a total whose allowed magnitudes run from the difference to the sum of the parts in integer steps. The change from the uncoupled product basis to the coupled total-angular-momentum basis is carried out with the lowering operator and orthogonality, and its matrix of overlaps is the table of Clebsch–Gordan coefficients. Two spin-halves split into a triplet and a singlet, the prototype for every composite spin.\n",{"path":15878,"title":15879,"module":15880,"summary":15881},"\u002Fquantum-mechanics\u002Fcentral-potentials\u002Fthe-schrodinger-equation-in-three-dimensions","The Schrödinger Equation in Three Dimensions","Central Potentials","A central potential depends only on the distance from a force center, so the three-dimensional Schrödinger equation separates in spherical coordinates. The angular factor is a spherical harmonic; the radial factor obeys a one-dimensional equation with an effective potential whose centrifugal barrier depends on the angular-momentum quantum number. The free particle and the spherical box fix the two limiting cases through the spherical Bessel functions.\n",{"path":15883,"title":15884,"module":15880,"summary":15885},"\u002Fquantum-mechanics\u002Fcentral-potentials\u002Fthe-hydrogen-atom","The Hydrogen Atom","The Coulomb potential turns the radial equation into one whose bound states exist only for a discrete set of energies. A power-series solution truncated to keep the wavefunction normalizable forces the principal quantum number, and the energy comes out proportional to minus one over its square, recovering the Rydberg spectrum. The bound states are the associated Laguerre functions times spherical harmonics, and their energy depends on the principal number alone, giving an n-squared degeneracy larger than rotational symmetry can explain.\n",{"path":15887,"title":15888,"module":15880,"summary":15889},"\u002Fquantum-mechanics\u002Fcentral-potentials\u002Fthe-isotropic-oscillator-and-hidden-symmetry","The Isotropic Oscillator and Hidden Symmetry","The three-dimensional isotropic harmonic oscillator solves in both Cartesian and spherical bases, and the two solutions must agree on the degeneracy of every level. That agreement, and the accidental degeneracy of hydrogen, both come from a symmetry larger than rotation: the oscillator carries an SU(3) invariance built from a conserved quadrupole tensor, and the Coulomb problem carries an SO(4) invariance built from the conserved Runge–Lenz vector. These hidden symmetries pin the degeneracies that rotational invariance alone leaves unexplained.\n",{"path":15891,"title":15892,"module":15893,"summary":15894},"\u002Fquantum-mechanics\u002Fspin\u002Fspin-half-pauli-matrices-and-stern-gerlach","Spin-½, the Pauli Matrices, and Stern–Gerlach","Spin","A silver atom passing through an inhomogeneous magnetic field splits into two beams, not a smear. That single fact fixes the internal angular momentum of the electron to a two-valued quantity with no spatial wavefunction. We build the two-dimensional spin space, the Pauli matrices and their algebra, the spinor for measurement along an arbitrary axis, and the sequential Stern–Gerlach filters that expose measurement disturbance.\n",{"path":15896,"title":15897,"module":15893,"summary":15898},"\u002Fquantum-mechanics\u002Fspin\u002Fspin-in-a-magnetic-field-precession-and-resonance","Spin in a Magnetic Field: Precession and Resonance","A spin coupled to a magnetic field is the simplest nontrivial quantum dynamics. A static field makes the spin expectation precess on a cone at the Larmor frequency while the energy levels split linearly. Adding a weak oscillating field and passing to the rotating frame produces Rabi oscillations and a resonance lineshape — the physics of NMR and ESR, and the driven qubit.\n",{"path":15900,"title":15901,"module":15893,"summary":15902},"\u002Fquantum-mechanics\u002Fspin\u002Ftwo-level-systems-and-the-bloch-sphere","Two-Level Systems and the Bloch Sphere","Every two-state quantum system is a spin-½ in disguise. Its Hamiltonian is an effective magnetic field, its pure states are points on the Bloch sphere, and its unitary evolution is a rigid rotation of that sphere. The same structure produces avoided level crossings, the ammonia inversion doublet and its maser, and the qubit.\n",{"path":15904,"title":15905,"module":15906,"summary":15907},"\u002Fquantum-mechanics\u002Fidentical-particles\u002Fidentical-particles-and-exchange-symmetry","Identical Particles and Exchange Symmetry","Identical Particles","Two electrons carry no label that distinguishes one from the other, and that bare fact reshapes the state space. The exchange operator that swaps particle labels commutes with any Hamiltonian built from identical particles, so its eigenvalue is conserved, and nature admits only its two extremes: totally symmetric states for bosons and totally antisymmetric states for fermions. The antisymmetry forces a statistical correlation, the exchange \"force,\" that keeps fermions apart and draws bosons together even with no interaction between them.\n",{"path":15909,"title":15910,"module":15906,"summary":15911},"\u002Fquantum-mechanics\u002Fidentical-particles\u002Fthe-pauli-principle-atoms-and-the-periodic-table","The Pauli Principle, Atoms, and the Periodic Table","Antisymmetry packaged as a Slater determinant turns the exclusion principle into an operating rule for building atoms. Helium shows the machinery in full: the electron-electron repulsion splits into a direct Coulomb integral and an exchange integral, and the exchange term alone pushes the spin-triplet (orthohelium) below the spin-singlet (parahelium) with no magnetic interaction in sight. Screening, the aufbau order, and Hund's rules then assemble the whole periodic table from the same antisymmetry.\n",{"path":15913,"title":15914,"module":15915,"summary":15916},"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Ftime-independent-perturbation-theory","Time-Independent Perturbation Theory","Approximation Methods for Bound States","Almost no realistic Hamiltonian can be solved exactly. Perturbation theory treats a hard Hamiltonian as a solvable one plus a small correction and expands the eigenvalues and eigenstates in powers of that correction. We derive the first- and second-order energy shifts and the first-order state correction for a nondegenerate level, expose the small-denominator failure that degeneracy forces, and fix it by diagonalizing the perturbation inside the degenerate subspace to find the \"good\" zeroth-order states.\n",{"path":15918,"title":15919,"module":15915,"summary":15920},"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Ffine-structure-and-the-real-hydrogen-atom","Fine Structure and the Real Hydrogen Atom","The Bohr spectrum is only the leading term. Two relativistic corrections of order alpha-squared — the relativistic kinetic-energy correction and spin–orbit coupling, joined by the Darwin term for s states — split the hydrogen levels into fine structure that depends on the total angular momentum j. We derive each shift as a first-order perturbation, combine them into a formula depending only on n and j, and continue down the energy ladder to the Lamb shift and the hyperfine 21 cm line.\n",{"path":15922,"title":15923,"module":15915,"summary":15924},"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Fthe-zeeman-and-stark-effects","The Zeeman and Stark Effects","An atom in an external field is a perturbation problem whose good basis depends on which interaction wins. A magnetic field competes with the internal spin–orbit coupling: the weak-field limit gives the anomalous Zeeman splitting set by the Landé g-factor, the strong-field limit gives the Paschen–Back pattern in the uncoupled basis, and the intermediate regime is a matrix diagonalization. An electric field gives a quadratic shift for the nondegenerate ground state and a linear splitting for the degenerate n = 2 level.\n",{"path":15926,"title":15927,"module":15915,"summary":15928},"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Fthe-variational-method","The Variational Method","The expectation of the Hamiltonian in any trial state is an upper bound on the true ground-state energy. Minimizing that expectation over a parametrized family of trial functions turns the ground-state problem into ordinary calculus and needs no small parameter. We prove the bound, apply it to the helium atom with a screened effective charge, use a two-center trial to predict binding in the hydrogen molecular ion, and extend the method to excited states through orthogonality.\n",{"path":15930,"title":15931,"module":15915,"summary":15932},"\u002Fquantum-mechanics\u002Fapproximation-methods\u002Fthe-wkb-approximation","The WKB Approximation","When the potential varies slowly on the scale of the de Broglie wavelength, the wavefunction is locally a plane wave with a position-dependent wavelength. This semiclassical picture builds the wavefunction from the classical momentum, breaks down at the turning points where the momentum vanishes, and is repaired there by connection formulas. The result recovers the Bohr–Sommerfeld quantization rule with its half-integer correction and gives the exponential tunneling rate through a smooth barrier, the Gamow factor.\n",{"path":15934,"title":15935,"module":6,"summary":6},"\u002Fquantum-mechanics","Quantum Mechanics",{"path":15937,"title":15938,"module":15939,"summary":15940},"\u002Freal-analysis\u002Ffoundations\u002Fsets-logic-functions","Sets, Logic, and Functions","Foundations and the Real Number System","The working language of analysis: quantifiers and the proof patterns (contrapositive, contradiction, induction), sets and their operations, relations and equivalence classes, and functions with their images, injections, surjections, and bijections. Cardinality is measured by bijection, and Cantor's theorem that no set surjects onto its power set forces uncountable sets to exist.\n",{"path":15942,"title":15943,"module":15939,"summary":15944},"\u002Freal-analysis\u002Ffoundations\u002Fordered-fields-completeness","Ordered Fields and the Completeness Axiom","The real numbers are the unique ordered field with the least-upper-bound property. The field and order axioms, the exact failure of the rationals (no supremum for the set of rationals below √2), and completeness as the defining axiom of ℝ lead to the first consequences: the existence of √2, the Archimedean property, and the density of ℚ in ℝ.\n",{"path":15946,"title":15947,"module":15939,"summary":15948},"\u002Freal-analysis\u002Ffoundations\u002Fabsolute-value-bounds","Absolute Value, Bounded Sets, and Inequalities","The absolute value turns the order on ℝ into a notion of distance, with the triangle inequality as the estimate underlying most later proofs. Covered: its algebra, the triangle and reverse-triangle inequalities, and the extension of the sup\u002Finf vocabulary from sets to bounded functions.\n",{"path":15950,"title":15951,"module":15939,"summary":15952},"\u002Freal-analysis\u002Ffoundations\u002Fintervals-uncountability","Intervals, Uncountability, and Decimals","Intervals are classified, and ℝ is proved uncountable two ways: a nested-interval construction and the decimal diagonal argument. Decimal expansions are built as suprema of truncations, which pins the source of their non-uniqueness (the 0.4999… equals 0.5000… identity) and the identification of the rationals with the eventually-repeating expansions. The middle-thirds Cantor set is an uncountable set of measure zero.\n",{"path":15954,"title":15955,"module":15956,"summary":15957},"\u002Freal-analysis\u002Fsequences-series\u002Fsequences-limits","Sequences and Their Limits","Sequences and Series","A sequence is a function on the natural numbers; it converges to a limit when its terms eventually stay within any prescribed tolerance of that number. The epsilon-M definition fixes the order of the quantifiers, and from it the limit is unique, every convergent sequence is bounded, and only the tail matters. Divergence to plus or minus infinity records terms that outgrow every bound.\n",{"path":15959,"title":15960,"module":15956,"summary":15961},"\u002Freal-analysis\u002Fsequences-series\u002Flimit-laws-monotone","Limit Laws and Monotone Convergence","Limits commute with sums, products, quotients, roots, and absolute values and preserve non-strict inequalities, so a limit can be assembled from the limits of its parts without returning to epsilon and M. The squeeze lemma transfers a limit through two envelopes; the monotone convergence theorem produces a limit from boundedness alone; and the ratio test settles the geometric and factorial standard limits.\n",{"path":15963,"title":15964,"module":15956,"summary":15965},"\u002Freal-analysis\u002Fsequences-series\u002Flimsup-bolzano-weierstrass","Subsequences, Limit Superior, and Bolzano–Weierstrass","A bounded sequence need not converge, but it always has convergent subsequences, and its terms cluster between two extreme values. The limit superior and inferior are the limits of the tail suprema and infima; they always exist for a bounded sequence, coincide exactly when it converges, and are its largest and smallest subsequential limits. Bolzano–Weierstrass extracts a convergent subsequence from boundedness alone.\n",{"path":15967,"title":15968,"module":15956,"summary":15969},"\u002Freal-analysis\u002Fsequences-series\u002Fcauchy-completeness","Cauchy Sequences and the Completeness of the Reals","The Cauchy criterion tests convergence without knowing the limit: a sequence converges exactly when its terms eventually all lie within any tolerance of one another. Cauchy sequences are bounded, in the reals Cauchy and convergent are equivalent, and this completeness property is interchangeable with the least-upper-bound axiom — the single feature that separates the real line from the rationals.\n",{"path":15971,"title":15972,"module":15956,"summary":15973},"\u002Freal-analysis\u002Fsequences-series\u002Fseries-convergence","Series and Convergence Tests","A series converges when its sequence of partial sums does, so every fact about sequences transfers. Geometric and telescoping series sum in closed form; the n-th term test rejects series whose terms miss zero, though the harmonic series shows the converse fails; and the comparison test against the geometric and p-series benchmarks settles most nonnegative-term series.\n",{"path":15975,"title":15976,"module":15956,"summary":15977},"\u002Freal-analysis\u002Fsequences-series\u002Fabsolute-conditional-rearrangement","Absolute Convergence, the Ratio and Root Tests, and Rearrangements","Absolute convergence is the strong form of convergence that permits free manipulation; conditional convergence is fragile. Absolute convergence implies convergence, and the ratio and root tests detect it by comparison with the geometric series. The alternating series test supplies conditionally convergent series, Riemann's theorem rearranges any of them to any sum, and Mertens' theorem multiplies series when at least one converges absolutely.\n",{"path":15979,"title":15980,"module":15981,"summary":15982},"\u002Freal-analysis\u002Fmetric-spaces\u002Fmetric-spaces-norms","Metric Spaces, Norms, and Examples","Metric Spaces and Topology","A metric is a function $d(x,y)$ obeying four axioms: nonnegativity, identity of indiscernibles, symmetry, and the triangle inequality. The Euclidean, taxicab, sup, discrete, and great-circle metrics all qualify, as does the sup metric on $C[a,b]$. Every norm induces a metric, and strongly equivalent metrics share the same open sets.\n",{"path":15984,"title":15985,"module":15981,"summary":15986},"\u002Freal-analysis\u002Fmetric-spaces\u002Fopen-closed-sets","Open and Closed Sets, Interior, Closure","Open sets are those in which every point has room to move; closed sets are their complements. From the single ball construction come the topology axioms (arbitrary unions, finite intersections), the interior, closure, and boundary of a set, and the fact that openness is always relative to the ambient space.\n",{"path":15988,"title":15989,"module":15981,"summary":15990},"\u002Freal-analysis\u002Fmetric-spaces\u002Fconvergence-completeness","Convergence, Cauchy Sequences, and Completeness","The $\\varepsilon$-$N$ definition of a limit transfers verbatim to any metric space once $|x-y|$ is replaced by $d(x,y)$. Convergent sequences characterize closed sets and closures; Cauchy sequences and completeness capture spaces with no missing limits, with $\\mathbb{R}^n$ and $C[a,b]$ complete and $\\mathbb{Q}$ and $(0,1]$ not.\n",{"path":15992,"title":15993,"module":15981,"summary":15994},"\u002Freal-analysis\u002Fmetric-spaces\u002Fcompactness","Compactness","A set is compact if every open cover has a finite subcover. In a metric space this is equivalent to sequential compactness and to being complete and totally bounded. Compact sets are closed and bounded; the Heine–Borel theorem gives the converse in $\\mathbb{R}^n$ but nowhere else in general.\n",{"path":15996,"title":15997,"module":15981,"summary":15998},"\u002Freal-analysis\u002Fmetric-spaces\u002Fconnectedness","Connectedness","A space is connected when it cannot be split into two nonempty open pieces. The connected subsets of $\\mathbb{R}$ are precisely the intervals, path- connectedness gives a constructive sufficient condition, and connectedness is a topological invariant preserved by continuous maps, the fact behind the intermediate value theorem.\n",{"path":16000,"title":16001,"module":14401,"summary":16002},"\u002Freal-analysis\u002Fcontinuity\u002Flimits-of-functions","Limits of Functions","The limit of a function at a point is an epsilon–delta condition pinning one value L as the target of f(x) as x approaches c, mirroring the sequence definition with distance replacing index. It is stated only at cluster points of the domain, is unique when it exists, and reduces to sequential limits through the Heine criterion. The algebra of limits and one-sided limits follow from that reduction.\n",{"path":16004,"title":16005,"module":14401,"summary":16006},"\u002Freal-analysis\u002Fcontinuity\u002Fcontinuous-functions","Continuous Functions","A function is continuous at c when its limit there equals its own value, lim f(x) = f(c). The epsilon–delta and sequential forms agree; sums, products, quotients, and compositions of continuous functions are continuous; and the failures split into jump, Dirichlet, popcorn, and removable types. The topological reading is that preimages of open sets are open.\n",{"path":16008,"title":16009,"module":14401,"summary":16010},"\u002Freal-analysis\u002Fcontinuity\u002Fevt-ivt","Extreme and Intermediate Value Theorems","On a closed bounded interval a continuous function attains an absolute maximum and minimum (the extreme value theorem, compactness preserved by continuity) and takes every value between its endpoint values (the intermediate value theorem, connectedness preserved). Both proofs run through Bolzano–Weierstrass and bisection, and yield root-finding, existence of k-th roots, and fixed-point theorems.\n",{"path":16012,"title":16013,"module":14401,"summary":16014},"\u002Freal-analysis\u002Fcontinuity\u002Funiform-continuity","Uniform Continuity","Uniform continuity strengthens continuity by demanding one delta that works at every point of the domain, not a delta re-chosen at each point. It separates x^2 on a compact interval from x^2 on the whole line and from 1\u002Fx near zero; continuity on a closed bounded interval is automatically uniform; uniformly continuous functions preserve Cauchy sequences and extend to endpoints; and Lipschitz continuity is the strongest of the three, through its secant-slope bound.\n",{"path":16016,"title":16017,"module":14401,"summary":16018},"\u002Freal-analysis\u002Fcontinuity\u002Fcontinuity-metric-spaces","Continuity on Metric Spaces","The epsilon–delta definition used only distances, so continuity transfers to maps between metric spaces by replacing absolute values with the two metrics. In this generality continuity still admits a sequential form, preserves compactness and connectedness, is uniform on a compact domain, and reads topologically as preimages of open sets being open, the formulation that defines homeomorphisms.\n",{"path":16020,"title":16021,"module":14401,"summary":16022},"\u002Freal-analysis\u002Fcontinuity\u002Flimits-infinity-monotone","Limits at Infinity and Monotone Functions","Treating infinity as a cluster point extends the epsilon–delta limit to x approaching plus or minus infinity, giving horizontal asymptotes and infinite limits. For monotone functions the one-sided limits always exist as suprema and infima, the discontinuities are jumps and at most countably many, the continuity is equivalent to the image being an interval, and a strictly monotone function always has a continuous inverse.\n",{"path":16024,"title":16025,"module":16026,"summary":16027},"\u002Freal-analysis\u002Fdifferentiation\u002Fthe-derivative","The Derivative","Differentiation","The derivative is the limit of the difference quotient, the slope the secant lines approach as the second point slides into the first. Differentiability forces continuity; linearity and the product, quotient, and chain rules follow from the definition; and a continuous function can fail to be differentiable, as the absolute value does at the origin.\n",{"path":16029,"title":16030,"module":16026,"summary":16031},"\u002Freal-analysis\u002Fdifferentiation\u002Fmean-value-theorem","The Mean Value Theorem","A relative extremum in the interior forces the derivative to vanish; Rolle's theorem and the mean value theorem turn that local fact into global control. The sign of the derivative fixes monotonicity, a bounded derivative yields a Lipschitz bound, and Darboux's theorem shows derivatives have the intermediate value property even where they are discontinuous.\n",{"path":16033,"title":16034,"module":16026,"summary":16035},"\u002Freal-analysis\u002Fdifferentiation\u002Ftaylors-theorem","Taylor's Theorem","Taylor's theorem generalizes the mean value theorem: an n-times differentiable function is matched near a point by a degree-n polynomial, with a Lagrange remainder that names the error exactly through one higher derivative. Iterating the mean value theorem proves it; the second-derivative test is the order-one case; and a smooth non-analytic bump separates a Taylor series from the function it fails to represent.\n",{"path":16037,"title":16038,"module":16026,"summary":16039},"\u002Freal-analysis\u002Fdifferentiation\u002Finverse-function-1d","The Inverse Function Theorem in One Variable","A nonzero derivative certifies a local inverse and fixes its slope. A strictly monotone differentiable function has a differentiable inverse whose derivative is the reciprocal of the original; the inverse function theorem removes the monotonicity hypothesis, and the reciprocal formula constructs nth roots and the logarithm's derivative, failing exactly where the derivative vanishes.\n",{"path":16041,"title":16042,"module":16043,"summary":16044},"\u002Freal-analysis\u002Friemann-integration\u002Fdarboux-integral","Partitions, Darboux Sums, and Integrability","The Riemann Integral","The Riemann integral is defined by trapping the area under a bounded function between under- and over-estimates. Partitions cut the domain into strips; lower and upper Darboux sums bracket the area; refining a partition tightens the bracket. A function is integrable exactly when the bracket can be made arbitrarily thin, and the tagged Riemann-sum limit gives the same number.\n",{"path":16046,"title":16047,"module":16043,"summary":16048},"\u002Freal-analysis\u002Friemann-integration\u002Fintegrability-classes","Which Functions Are Integrable","The Cauchy criterion certifies whole classes of functions as integrable. Continuous functions are integrable because uniform continuity makes every oscillation cap small; monotone functions are integrable because their caps telescope to a single total jump; bounded functions with finitely many discontinuities are integrable by isolating the bad points. The Dirichlet function fails, and the Lebesgue criterion names the exact boundary.\n",{"path":16050,"title":16051,"module":16043,"summary":16052},"\u002Freal-analysis\u002Friemann-integration\u002Fproperties-of-the-integral","Properties of the Integral","The integral is a linear, order-preserving, additive operator on the integrable functions. It splits across subintervals, respects inequalities, bounds the size of a function by the integral of its absolute value, and preserves products. The mean value theorem for integrals identifies the integral with an attained average height on a fixed rectangle.\n",{"path":16054,"title":14456,"module":16043,"summary":16055},"\u002Freal-analysis\u002Friemann-integration\u002Ffundamental-theorem","The fundamental theorem ties the integral to the derivative in two forms. The evaluation form computes a definite integral from any antiderivative; the differentiation form shows the area function has derivative equal to the integrand at points of continuity. Together they make differentiation and integration inverse operations, and yield integration by parts and change of variables.\n",{"path":16057,"title":16058,"module":16043,"summary":16059},"\u002Freal-analysis\u002Friemann-integration\u002Flog-exp-improper","The Logarithm, Exponential, and Improper Integrals","The integral defines transcendental functions. The logarithm is the area under 1\u002Ft, the exponential is its inverse, and their calculus properties follow from the fundamental theorem. Improper integrals extend integration to unbounded intervals and unbounded integrands as limits of proper integrals, with a p-test, a comparison test, absolute versus conditional convergence, and the integral test linking integrals to series.\n",{"path":16061,"title":16062,"module":16063,"summary":16064},"\u002Freal-analysis\u002Ffunction-sequences\u002Fpointwise-uniform-convergence","Pointwise and Uniform Convergence","Sequences and Series of Functions","A sequence of functions has two natural notions of limit. Pointwise convergence fixes each input and takes the limit of numbers; uniform convergence demands one rate that works for every input at once. The uniform norm turns the second into a statement about a single sequence of numbers, and the uniform Cauchy criterion and the Weierstrass M-test let us certify it.\n",{"path":16066,"title":16067,"module":16063,"summary":16068},"\u002Freal-analysis\u002Ffunction-sequences\u002Finterchange-of-limits","Interchange of Limits: Continuity, Integration, Differentiation","Passing to a limit inside a continuity statement, an integral, or a derivative is an interchange of two limits, and the two limits do not always commute. Uniform convergence licenses the first two swaps: the uniform limit of continuous functions is continuous, and the limit of the integrals is the integral of the limit. Differentiation needs uniform convergence of the derivatives, and counterexamples show why each hypothesis is required.\n",{"path":16070,"title":16071,"module":16063,"summary":16072},"\u002Freal-analysis\u002Ffunction-sequences\u002Fpower-series-weierstrass","Power Series and the Weierstrass Approximation Theorem","A power series converges uniformly on every closed subinterval inside its radius of convergence, together with all of its derivatives. That makes it continuous, differentiable, and integrable term by term, so a power series defines an infinitely differentiable function. The Weierstrass approximation theorem then shows that polynomials come uniformly close to any continuous function on a closed bounded interval.\n",{"path":16074,"title":16075,"module":16063,"summary":16076},"\u002Freal-analysis\u002Ffunction-sequences\u002Fpicard-ode","Picard's Existence and Uniqueness Theorem","The Banach fixed-point theorem says a contraction of a complete metric space has exactly one fixed point, found by iterating from any start. Applied to the space of continuous functions with the uniform norm, it proves Picard's theorem: a first-order differential equation with a Lipschitz right-hand side has a unique local solution. Picard iteration constructs that solution explicitly, and worked examples show the Lipschitz condition is not optional.\n",{"path":16078,"title":16079,"module":16080,"summary":16081},"\u002Freal-analysis\u002Fseveral-variables\u002Fdifferentiability-rn","The Derivative of a Map ℝⁿ → ℝᵐ","Functions of Several Variables (Introduction)","The derivative of a map between Euclidean spaces is the linear transformation of vanishing relative error, unique when it exists and represented in coordinates by the Jacobian matrix of partial derivatives. Differentiability forces continuity through a local Lipschitz bound. Existence of the partial derivatives alone does not suffice; continuity of the partials does.\n",{"path":16083,"title":16084,"module":16080,"summary":16085},"\u002Freal-analysis\u002Fseveral-variables\u002Fgradient-chain-rule","Directional Derivatives, the Gradient, and the Chain Rule","The directional derivative measures the rate of change of a scalar field along a chosen heading and equals the derivative applied to that direction. The gradient collects these into a vector that points along steepest ascent and sits orthogonal to level sets. The chain rule composes derivatives by multiplying Jacobians, and a mean value theorem holds for scalar fields but fails for vector-valued maps.\n",{"path":16087,"title":16088,"module":16080,"summary":16089},"\u002Freal-analysis\u002Fseveral-variables\u002Fhigher-derivatives-taylor-extrema","Higher Derivatives, Taylor's Theorem, and Extrema","Iterating the derivative gives a symmetric second derivative, the Hessian, whose mixed partials agree when they are continuous. Taylor's theorem expands a smooth map to any order with a Lagrange-type remainder, and at a critical point the definiteness of the Hessian decides between a local minimum, a local maximum, and a saddle.\n",{"path":16091,"title":16092,"module":16080,"summary":16093},"\u002Freal-analysis\u002Fseveral-variables\u002Finverse-implicit-theorems","The Inverse and Implicit Function Theorems","A nonlinear map with a nonsingular Jacobian is locally invertible, with the inverse's derivative given by the inverse matrix. The contraction mapping principle supplies the local inverse; the implicit function theorem then solves a system for some variables in terms of the rest whenever the relevant Jacobian block is invertible. Worked coordinate changes show both theorems in use.\n",{"path":16095,"title":16096,"module":16080,"summary":16097},"\u002Freal-analysis\u002Fseveral-variables\u002Fmultiple-integrals","Multiple Integrals","The Riemann integral of a bounded function over a closed rectangle in Euclidean space is built from Darboux upper and lower sums on a grid of subrectangles, with the same squeeze criterion that governs the one-variable integral. Continuous integrands are integrable, and a set of content zero can be ignored. Fubini's theorem evaluates a multiple integral as an iterated one in either order, and the indicator trick extends the theory to regions bounded by curves.\n",{"path":16099,"title":16100,"module":6,"summary":6},"\u002Freal-analysis","Real Analysis",{"path":16102,"title":16103,"module":14054,"summary":16104},"\u002Fabstract-algebra\u002Ffoundations\u002Fsets-functions-relations","Sets, Functions, and Equivalence Relations","Algebra is built on three prior notions: the set, the map between sets, and the equivalence relation that reorganizes a set into disjoint classes. Sets, maps (injective, surjective, bijective), fibers and preimages, and the correspondence between equivalence relations and partitions — the one structural fact reused in every later quotient construction.\n",{"path":16106,"title":16107,"module":14054,"summary":16108},"\u002Fabstract-algebra\u002Ffoundations\u002Fintegers-and-modular-arithmetic","The Integers and Modular Arithmetic","The integers carry the template every ring later imitates: well-ordering drives induction, induction drives the division algorithm, and division drives the Euclidean algorithm, gcd, Bézout's identity, and unique factorization into primes. Quotienting by congruence mod n builds the first finite arithmetic, Z\u002FnZ, whose invertible elements form the group of units.\n",{"path":16110,"title":16111,"module":16112,"summary":16113},"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fgroup-axioms-and-first-examples","Group Axioms and First Examples","Groups and Symmetry","A group is a set with one associative operation that has an identity and inverses. We state the axioms, prove that the identity, inverses, and cancellation behave as expected, define the order of a group and of an element, and catalogue the running examples: the integers, the additive group of residues mod n, and the multiplicative group of units mod n.\n",{"path":16115,"title":16116,"module":16112,"summary":16117},"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fdihedral-and-symmetric-groups","Dihedral and Symmetric Groups","The dihedral group D_{2n} is the symmetries of a regular n-gon, generated by a rotation r and a reflection s subject to three relations. The symmetric group S_n is all permutations of n objects, written in cycle notation. Orders, generators and relations, cycle decomposition, the order of a permutation from its cycle type, and the parity that splits S_n in half.\n",{"path":16119,"title":16120,"module":16112,"summary":16121},"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fmatrix-and-quaternion-groups","Matrix and Quaternion Groups","Invertible matrices over a field form the general linear group GL_n(F), with the determinant-one matrices as the subgroup SL_n(F). Over a finite field the order of GL_n(F) has a clean product formula. The quaternion group Q_8 is a second small nonabelian group, distinct from the dihedral group of the same order; its multiplication and subgroup structure sharpen the contrast between the two.\n",{"path":16123,"title":16124,"module":16112,"summary":16125},"\u002Fabstract-algebra\u002Fgroups-and-symmetry\u002Fhomomorphisms-and-group-actions","Homomorphisms, Isomorphisms, and Actions","A homomorphism is a map between groups that respects the operation; an isomorphism is a bijective one, making two groups the same up to relabeling. The kernel and image measure how far a homomorphism is from injective and surjective. A group action realizes a group as permutations of a set, and actions correspond exactly to homomorphisms into a symmetric group, with orbits and stabilizers as the first tools for counting.\n",{"path":16127,"title":16128,"module":16129,"summary":16130},"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fsubgroups-and-substructures","Subgroups and Their Substructures","Subgroups and Quotients","A subgroup is a subset that is a group under the inherited operation. One test decides it: nonempty and closed under the map $(x,y) \\mapsto xy^{-1}$. From an arbitrary subset $A$ we build the centralizer, normalizer, and center, and from an action the stabilizer and kernel, all of them subgroups nested in a fixed chain inside $G$.\n",{"path":16132,"title":16133,"module":16129,"summary":16134},"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fcyclic-groups","Cyclic Groups","A cyclic group is generated by one element. Two facts organize the whole theory: the order of an element equals the order of the subgroup it generates, and cyclic groups of equal order are isomorphic, so $\\mathbb{Z}$ and $\\mathbb{Z}\u002Fn\\mathbb{Z}$ are the only ones. From there the generators ($\\varphi(n)$ of them), the subgroups (one per divisor of $n$), and a fast exponentiation algorithm all follow.\n",{"path":16136,"title":16137,"module":16129,"summary":16138},"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fgeneration-and-subgroup-lattices","Generation and the Lattice of Subgroups","The subgroup generated by a subset $A$ is the smallest subgroup containing it, described top-down as an intersection and bottom-up as the set of words in $A$ and its inverses. Collecting all subgroups and ordering them by containment produces the subgroup lattice, whose Hasse diagram shows the joins, meets, and containment relations among all subgroups.\n",{"path":16140,"title":16141,"module":16129,"summary":16142},"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fcosets-lagrange-and-normal-subgroups","Cosets, Lagrange, and Normal Subgroups","The left cosets of a subgroup partition a group into equal-sized blocks, so the order of a subgroup divides the order of the group: Lagrange's theorem. When the blocks can be multiplied consistently — exactly when the subgroup is normal — they form the quotient group $G\u002FN$. Fermat's and Euler's theorems fall out as index computations.\n",{"path":16144,"title":16145,"module":16129,"summary":16146},"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fisomorphism-theorems","The Isomorphism Theorems","Four theorems relate homomorphisms, quotients, and subgroup lattices. The first identifies the image of a homomorphism with the quotient by its kernel; the second and third compute quotients built from two subgroups and quotients of quotients; the fourth matches the subgroups of $G\u002FN$ with the subgroups of $G$ lying above $N$. Together they make quotient groups computable.\n",{"path":16148,"title":16149,"module":16129,"summary":16150},"\u002Fabstract-algebra\u002Fsubgroups-and-quotients\u002Fcomposition-series-and-the-alternating-group","Composition Series and the Alternating Group","A composition series breaks a finite group into simple quotient factors, and Jordan-Hölder says those factors are unique up to order. This turns classification into two problems: list the simple groups, and describe how to reassemble them. The sign homomorphism splits $S_n$ into even and odd permutations, defining the alternating group $A_n$, simple for $n \\ge 5$.\n",{"path":16152,"title":16153,"module":16154,"summary":16155},"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Factions-and-cayleys-theorem","Actions, Orbits, and Cayley's Theorem","Group Actions and Sylow Theory","A group action turns abstract elements into permutations of a set. The action splits the set into orbits, and the orbit-stabilizer theorem ties each orbit's size to the index of a stabilizer. Applied to a group acting on itself by left multiplication, this gives Cayley's theorem: every group is a group of permutations.\n",{"path":16157,"title":16158,"module":16154,"summary":16159},"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Fconjugation-and-the-class-equation","Conjugation and the Class Equation","A group acts on itself by conjugation, and the orbits are the conjugacy classes. Orbit-stabilizer turns the resulting partition into the class equation, which forces every group of prime-power order to have a nontrivial center. Conjugacy in the symmetric group is cycle type, and Burnside's lemma counts orbits by averaging fixed points.\n",{"path":16161,"title":16162,"module":16154,"summary":16163},"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Fsylow-theorems","The Sylow Theorems","Lagrange's theorem forbids subgroups whose order fails to divide the group order; Sylow's theorems supply a partial converse for prime powers. A Sylow p-subgroup always exists, all of them are conjugate, and their count satisfies two congruence-and-divisibility constraints tight enough to prove many groups non-simple from their order alone.\n",{"path":16165,"title":16166,"module":16154,"summary":16167},"\u002Fabstract-algebra\u002Fgroup-actions-and-sylow\u002Fautomorphisms-and-simple-groups","Automorphisms and Simplicity of Aₙ","Conjugation makes a group act on itself and on its normal subgroups by automorphisms, giving the inner automorphism group G\u002FZ(G) and the embedding of N(H)\u002FC(H) into Aut(H). Characteristic subgroups are those every automorphism fixes, and the automorphism group of a cyclic group is its unit group. The lesson closes by proving the alternating group Aₙ is simple for n ≥ 5.\n",{"path":16169,"title":16170,"module":16171,"summary":16172},"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fdirect-products-and-finite-abelian-groups","Direct Products and Finite Abelian Groups","Products and Group Structure","The direct product assembles a larger group from componentwise copies of smaller ones, and a recognition theorem reverses the process when two normal subgroups meet trivially and span the group. The Fundamental Theorem of Finitely Generated Abelian Groups then classifies every such group by two equivalent invariants, invariant factors and elementary divisors.\n",{"path":16174,"title":16175,"module":16171,"summary":16176},"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fsemidirect-products","Semidirect Products","The semidirect product relaxes the direct product by requiring only one factor to be normal, with the other acting on it through a homomorphism into its automorphism group. This single twisting map lets abelian pieces assemble into non-abelian groups, realizes the dihedral groups as $\\mathbb{Z}_n \\rtimes \\mathbb{Z}_2$, and, with a recognition theorem, classifies groups of several small orders.\n",{"path":16178,"title":16179,"module":16171,"summary":16180},"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fnilpotent-and-solvable-groups","p-Groups, Nilpotent, and Solvable Groups","Finite p-groups have nontrivial center, and iterating the center upward builds the nilpotent groups, which decompose as the direct product of their Sylow subgroups. Iterating the commutator downward builds the solvable groups, whose factors are abelian. The chain cyclic, abelian, nilpotent, solvable orders these classes, and A_5 breaks the last link.\n",{"path":16182,"title":16183,"module":16171,"summary":16184},"\u002Fabstract-algebra\u002Fproducts-and-group-structure\u002Fclassifying-small-groups","Classifying Groups of Small Order","With Sylow's theorem to force normal subgroups, direct and semidirect products to assemble them, and presentations to name the result, every group up to order fifteen can be listed explicitly. Free groups make presentations precise: generators with no relations, from which any group is a quotient by the normal closure of its relations.\n",{"path":16186,"title":16187,"module":16188,"summary":16189},"\u002Fabstract-algebra\u002Fring-theory\u002Frings-definitions-and-examples","Rings: Definitions and Examples","Ring Theory","A ring carries two operations: an abelian group under addition and an associative multiplication linked by the distributive laws. The named special cases — commutative rings, integral domains, division rings, and fields — differ only in how their multiplication behaves. Standard examples include quadratic integer rings, polynomial rings, matrix rings, and group rings.\n",{"path":16191,"title":16192,"module":16188,"summary":16193},"\u002Fabstract-algebra\u002Fring-theory\u002Fideals-quotients-and-homomorphisms","Ideals, Quotient Rings, and Homomorphisms","Ring homomorphisms have kernels that absorb multiplication; such subsets are ideals, and every ideal is the kernel of the projection onto a quotient ring. The quotient construction yields the ring isomorphism theorems and classifies ideals by their quotients: R\u002FI is a field exactly when I is maximal, an integral domain exactly when I is prime.\n",{"path":16195,"title":16196,"module":16188,"summary":16197},"\u002Fabstract-algebra\u002Fring-theory\u002Ffractions-and-the-chinese-remainder-theorem","Fields of Fractions and the CRT","Rings of fractions invert a multiplicatively closed set, enlarging an integral domain into its field of fractions the way Z becomes Q. The Chinese Remainder Theorem splits a quotient by comaximal ideals into a direct product, generalizing Z\u002FmnZ ≅ Z\u002FmZ × Z\u002FnZ and explaining why the Euler function is multiplicative.\n",{"path":16199,"title":16200,"module":16201,"summary":16202},"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Feuclidean-domains-pids-ufds","Euclidean Domains, PIDs, and UFDs","Factorization and Polynomial Rings","Three classes of integral domain, ordered by how much of elementary arithmetic survives: Euclidean domains carry a division algorithm, principal ideal domains make every ideal a single multiple, and unique factorization domains factor every element into irreducibles in one way. We prove the chain ED implies PID implies UFD, the classes are separated by explicit counterexamples, and irreducible and prime coincide exactly in a UFD.\n",{"path":16204,"title":16205,"module":16201,"summary":16206},"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Fpolynomial-rings-over-fields","Polynomial Rings over Fields","When the coefficients form a field, polynomial long division works exactly as it does over the rationals, and it works with a unique quotient and remainder. That single fact makes F[x] a Euclidean domain, hence a PID and a UFD: every ideal is the multiples of one polynomial, roots correspond to linear factors, and F[x]\u002F(f) is a field precisely when f is irreducible.\n",{"path":16208,"title":16209,"module":16201,"summary":16210},"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Fgauss-lemma-and-unique-factorization","Gauss's Lemma and Unique Factorization","A UFD is not a field, so its polynomial ring is not a PID — yet unique factorization survives the passage from R to R[x]. Gauss's lemma supplies the passage: a polynomial that factors over the fraction field already factors over R, once content is factored out. This gives the theorem that R[x] is a UFD whenever R is, so Z[x] and Q[x,y] factor uniquely even though neither is a PID.\n",{"path":16212,"title":16213,"module":16201,"summary":16214},"\u002Fabstract-algebra\u002Ffactorization-and-polynomials\u002Firreducibility-criteria-and-groebner","Irreducibility Criteria and Gröbner Bases","Deciding whether a given polynomial is irreducible, and computing in multivariate polynomial rings. In one variable: the rational root test, reduction modulo a prime, and Eisenstein's criterion. In several variables, where division fails, a monomial order gives leading terms, a Gröbner basis restores a well-defined remainder, and Buchberger's algorithm computes it.\n",{"path":16216,"title":16217,"module":16218,"summary":16219},"\u002Fabstract-algebra\u002Fmodule-theory\u002Fintroduction-to-modules","Introduction to Modules","Module Theory","A module is an abelian group on which a ring acts, generalizing both vector spaces (when the ring is a field) and abelian groups (when the ring is the integers). Submodules, homomorphisms, quotients, and the isomorphism theorems carry over from groups, and an F[x]-module is the same datum as a vector space with a chosen linear operator — the correspondence behind the canonical forms.\n",{"path":16221,"title":16222,"module":16218,"summary":16223},"\u002Fabstract-algebra\u002Fmodule-theory\u002Ffree-modules-and-direct-sums","Generation, Direct Sums, and Free Modules","A generating set spans a module by R-linear combinations; a direct sum decomposes it into independent pieces; a free module has a basis and the universal property that a homomorphism is determined by arbitrary values on that basis. Rank is well defined over a commutative ring, torsion blocks a basis, and every module is a quotient of a free one — a presentation by generators and relations.\n",{"path":16225,"title":16226,"module":16218,"summary":16227},"\u002Fabstract-algebra\u002Fmodule-theory\u002Ftensor-products-and-exact-sequences","Tensor Products and Exact Sequences","The tensor product builds a module in which elements of two modules can be multiplied, characterized by a universal property turning bilinear maps into linear ones; extension of scalars is its guiding case. Exact sequences track how a module is assembled from a submodule and a quotient, when that assembly splits, and which modules — projective, injective, flat — make the Hom and tensor functors preserve exactness.\n",{"path":16229,"title":16230,"module":16218,"summary":16231},"\u002Fabstract-algebra\u002Fmodule-theory\u002Fvector-spaces-and-linear-maps","Vector Spaces and Linear Maps","A vector space is a module over a field, and the field hypothesis removes every pathology a general module can have: every vector space is free, so it has a basis, a well-defined dimension, and a coordinate isomorphism with F^n. Linear maps become matrices, change of basis becomes similarity, every space pairs with a dual of the same dimension, and the determinant is the unique alternating multilinear normalized form.\n",{"path":16233,"title":16234,"module":16235,"summary":16236},"\u002Fabstract-algebra\u002Fmodules-over-pids\u002Fstructure-theorem-over-pids","The Structure Theorem for Modules over a PID","Modules over PIDs and Canonical Forms","Every finitely generated module over a principal ideal domain splits as a free part plus a direct sum of cyclic torsion pieces, in two canonical ways: invariant factors, tied together by a divisibility chain, and elementary divisors, one prime power at a time. Existence follows from the stacked-basis theorem, both lists are unique, and the case $R = \\mathbb{Z}$ is the classification of finitely generated abelian groups.\n",{"path":16238,"title":16239,"module":16235,"summary":16240},"\u002Fabstract-algebra\u002Fmodules-over-pids\u002Frational-canonical-form","Rational Canonical Form","A linear operator turns its vector space into a module over the polynomial ring $F[x]$, with $x$ acting as the operator. The structure theorem's invariant factors then become polynomials, each cyclic summand becomes a companion matrix, and the block-diagonal assembly is the rational canonical form. It is unique, it is computed inside the base field, and two matrices are similar exactly when their rational canonical forms agree.\n",{"path":16242,"title":16243,"module":16235,"summary":16244},"\u002Fabstract-algebra\u002Fmodules-over-pids\u002Fjordan-canonical-form","Jordan Canonical Form","When the base field contains all the eigenvalues, the elementary divisors of an operator are powers of linear polynomials, and each cyclic summand becomes a Jordan block: an eigenvalue on the diagonal with ones just above it. Stacking the blocks gives the Jordan canonical form, unique up to reordering, as close to diagonal as the operator allows. Diagonalizability reads off the minimal polynomial, and the block sizes are counted by ranks of powers of the operator minus the eigenvalue.\n",{"path":16246,"title":16247,"module":16248,"summary":16249},"\u002Fabstract-algebra\u002Ffield-theory\u002Ffield-extensions-and-algebraic-elements","Field Extensions and Algebraic Elements","Field Theory","A field extension makes a larger field K into a vector space over a smaller field F, and its degree [K:F] is that dimension. Adjoining a root of an irreducible polynomial builds a simple extension F(α) isomorphic to F[x]\u002F(m), whose degree is the degree of the minimal polynomial. The tower law makes these degrees multiply, which turns algebra over fields into bookkeeping with integers.\n",{"path":16251,"title":16252,"module":16248,"summary":16253},"\u002Fabstract-algebra\u002Ffield-theory\u002Fstraightedge-and-compass-constructions","Straightedge-and-Compass Constructions","The lengths a straightedge and compass can build from a unit form a field closed under square roots, and every constructible number lies in a tower of quadratic extensions. So its degree over the rationals is a power of two. That single obstruction settles three problems the Greeks left open: doubling the cube, trisecting a general angle, and squaring the circle are all impossible.\n",{"path":16255,"title":16256,"module":16248,"summary":16257},"\u002Fabstract-algebra\u002Ffield-theory\u002Fsplitting-fields-and-algebraic-closure","Splitting Fields and Algebraic Closure","The splitting field of a polynomial is the smallest extension in which it factors into linear pieces, obtained by adjoining all its roots. Every polynomial has one, its degree is at most n factorial, and any two splitting fields are isomorphic. Pushing this to all polynomials at once gives the algebraic closure, a field in which every polynomial splits and which is unique up to isomorphism.\n",{"path":16259,"title":16260,"module":16248,"summary":16261},"\u002Fabstract-algebra\u002Ffield-theory\u002Fseparable-and-cyclotomic-extensions","Separable Extensions and Cyclotomic Fields","A polynomial is separable when its roots are distinct, detected by whether it shares a factor with its formal derivative. Over perfect fields — characteristic zero and finite fields — every irreducible is separable, and the existence and uniqueness of the finite fields follow. Cyclotomic polynomials package the roots of unity by order, are irreducible over the rationals, and give the cyclotomic field its degree phi(n).\n",{"path":16263,"title":16264,"module":16265,"summary":16266},"\u002Fabstract-algebra\u002Fgalois-theory\u002Fthe-galois-correspondence","The Galois Correspondence","Galois Theory","Galois theory attaches to a field extension its group of symmetries and shows that, for the right extensions, the subgroups of that group are in exact order-reversing correspondence with the intermediate fields. The automorphism group, Artin's theorem, the characterization of Galois extensions, and the Fundamental Theorem together turn questions about fields into questions about finite groups.\n",{"path":16268,"title":16269,"module":16265,"summary":16270},"\u002Fabstract-algebra\u002Fgalois-theory\u002Ffinite-fields","Finite Fields","Every finite field has prime-power order, is the splitting field of $x^{p^n} - x$, and is unique up to isomorphism. Its extension over the prime field is Galois with cyclic group generated by the Frobenius map $x \\mapsto x^p$, so the Galois correspondence reduces the subfield lattice to the divisor lattice of $n$. Möbius inversion counts the irreducible polynomials of each degree, and cyclic error-correcting codes are one application.\n",{"path":16272,"title":16273,"module":16265,"summary":16274},"\u002Fabstract-algebra\u002Fgalois-theory\u002Fcyclotomic-and-abelian-extensions","Cyclotomic and Abelian Extensions","The Galois group of the $n$th cyclotomic field over $\\mathbb{Q}$ is the unit group $(\\mathbb{Z}\u002Fn\\mathbb{Z})^\\times$, which makes cyclotomic fields the worked catalogue of abelian extensions of $\\mathbb{Q}$. The isomorphism identifies subfields with subgroups, realizes every finite abelian group as a Galois group over $\\mathbb{Q}$, and leads to Kronecker–Weber. Composites of Galois extensions and the primitive element theorem supply the machinery.\n",{"path":16276,"title":16277,"module":16265,"summary":16278},"\u002Fabstract-algebra\u002Fgalois-theory\u002Fgalois-groups-of-polynomials","Galois Groups of Polynomials","Ordering the roots of a separable polynomial embeds its Galois group in the symmetric group $S_n$, and the group is transitive exactly when the polynomial is irreducible. The discriminant decides membership in $A_n$; for cubics and quartics the resolvent cubic pins the group down; and reduction modulo a prime produces elements of prescribed cycle type, the standard tool for computing Galois groups over $\\mathbb{Q}$.\n",{"path":16280,"title":16281,"module":16265,"summary":16282},"\u002Fabstract-algebra\u002Fgalois-theory\u002Fsolvability-by-radicals-and-the-quintic","Solvability by Radicals and the Quintic","A polynomial is solvable by radicals exactly when its Galois group is solvable. Cyclic extensions are radical extensions once roots of unity are present, which turns a radical tower into a solvable subnormal series. Since $S_n$ is solvable only for $n \\le 4$, the general quintic has no radical formula, and an explicit quintic with Galois group $S_5$ has roots provably not expressible in radicals.\n",{"path":16284,"title":16285,"module":16286,"summary":16287},"\u002Fabstract-algebra\u002Fcapstone\u002Fcommutative-algebra-and-algebraic-geometry","A Glimpse of Commutative Algebra and Algebraic Geometry","Capstone: Where Algebra Goes Next","Commutative algebra reads geometry off the ring of polynomial functions. The dictionary runs through Noetherian rings and the ascending chain condition, Hilbert's Basis Theorem, affine algebraic sets and the two maps connecting ideals to zero sets, radicals, the Zariski topology, and Hilbert's Nullstellensatz, which over an algebraically closed field makes radical ideals and algebraic sets the same object.\n",{"path":16289,"title":16290,"module":16286,"summary":16291},"\u002Fabstract-algebra\u002Fcapstone\u002Frepresentation-and-character-theory","A Glimpse of Representation and Character Theory","Representation theory studies a group by the ways it can act linearly on a vector space. Representations are equivalent to modules over the group ring; Maschke's theorem gives complete reducibility, the Wedderburn consequences bound the irreducible degrees, and character theory reduces a representation to a trace invariant governed by the orthogonality relations and displayed in the character table of a small group.\n",{"path":16293,"title":16294,"module":6,"summary":6},"\u002Fabstract-algebra","Abstract Algebra",{"path":16296,"title":16297,"module":16298,"summary":16299},"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fatomic-spectra-rutherford","Atomic Spectra and Rutherford's Nucleus","Early Atomic Models and the Old Quantum Theory","Atoms emit light only at sharp, reproducible wavelengths, and by 1890 those wavelengths were captured by the Rydberg-Ritz formula. Neither empirical regularity had a mechanical explanation. Rutherford's alpha-scattering experiment supplied the missing structure: the atom's positive charge and nearly all its mass sit in a tiny central nucleus, with the electrons far outside.\n",{"path":16301,"title":16302,"module":16298,"summary":16303},"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fbohr-model-hydrogen","The Bohr Model of Hydrogen","Bohr grafted three quantum postulates onto Rutherford's nuclear atom: certain orbits do not radiate, radiation accompanies a jump between them, and quantization must match classical physics for large orbits. Quantizing the angular momentum fixes the orbit radii and energies, reproduces the Rydberg-Ritz formula, and predicts the Rydberg constant from fundamental constants alone.\n",{"path":16305,"title":16306,"module":16298,"summary":16307},"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fx-ray-spectra-franck-hertz","X-Ray Spectra and the Franck-Hertz Experiment","Two 1913-14 experiments confirmed the Bohr-Rutherford atom independently of optical spectra. Moseley found that the square root of a characteristic X-ray frequency is linear in atomic number, fixing Z as nuclear charge and ordering the periodic table. Franck and Hertz measured discrete atomic energy levels directly by scattering electrons through a mercury vapor.\n",{"path":16309,"title":16310,"module":16298,"summary":16311},"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fbohr-sommerfeld-old-quantum-theory","The Bohr-Sommerfeld Old Quantum Theory","Bohr fixed the hydrogen levels with a single quantum number by quantizing angular momentum. Sommerfeld replaced that ad hoc rule with a general prescription: quantize the action of each separable coordinate. The rule produces elliptical orbits, a second (azimuthal) quantum number, space quantization, and — once the relativistic mass variation is included — a fine-structure splitting that matches experiment to order alpha squared.\n",{"path":16313,"title":16314,"module":16298,"summary":16315},"\u002Fatomic-physics\u002Fearly-models-and-old-quantum-theory\u002Fold-quantum-theory-limits-wkb","Limits of the Old Quantum Theory and the WKB Bridge","The old quantum theory works only where the classical motion is separable into independent periodic coordinates. It fails for helium, forbids the correct zero angular momentum of the hydrogen ground state, and misses the half-integer in the oscillator and in molecular spectra. The WKB quantization condition, derived from the Schrodinger equation, is the modern descendant of the Sommerfeld rule and repairs the half-integer through the Maslov correction.\n",{"path":16317,"title":16318,"module":16319,"summary":16320},"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fschrodinger-3d-hydrogen","The Schrödinger Equation in Three Dimensions and Hydrogen","The Quantum Hydrogen Atom","Extending the Schrödinger equation to three dimensions and separating it in spherical coordinates produces three ordinary differential equations, one per coordinate. Their boundary conditions generate the quantum numbers n, ℓ, and mℓ, quantize the angular momentum to √(ℓ(ℓ+1))ℏ with projections mℏ, and fix the bound-state energies of hydrogen at −Z²(13.6 eV)\u002Fn².\n",{"path":16322,"title":16323,"module":16319,"summary":16324},"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fhydrogen-wave-functions","Hydrogen Wave Functions and Orbitals","The hydrogen wave functions factor into a radial part Rₙℓ(r) and an angular spherical harmonic Yℓm(θ,φ). Squaring gives the probability cloud; the radial distribution P(r) = r²|ψ|² peaks at the Bohr radius for the ground state and at the Bohr orbits for excited states. The angular part fixes the s, p, and d orbital shapes that govern chemical bonding.\n",{"path":16326,"title":16327,"module":16319,"summary":16328},"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fradial-equation-in-full","Solving the Radial Equation in Full","The hydrogen radial equation is solved from the differential equation up. The substitution u = rR turns it into a one-dimensional problem with a centrifugal barrier; matching the asymptotic behaviour at the origin and at infinity peels off the factors r^(ℓ+1) and e^(−r\u002Fna₀); a Frobenius series for the remainder must terminate, and that termination condition yields the quantization n ≥ ℓ+1 with E = −Z²Ry\u002Fn². The surviving polynomials are the associated Laguerre functions, whose degree n−ℓ−1 counts the radial nodes.\n",{"path":16330,"title":16331,"module":16319,"summary":16332},"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fsymmetry-degeneracy-runge-lenz","Accidental Degeneracy and the Runge-Lenz Symmetry","Hydrogen energies depend only on n, so states of different ℓ at the same n are degenerate. This is not a coincidence but the mark of a hidden symmetry: the quantum Runge-Lenz vector is conserved for the 1\u002Fr potential alone, and together with angular momentum it generates the group SO(4). The Casimir invariant of that group reproduces E = −Z²Ry\u002Fn² and its representations count the n² states. Any departure from 1\u002Fr breaks the symmetry and lifts the ℓ-degeneracy.\n",{"path":16334,"title":16335,"module":16319,"summary":16336},"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fexpectation-values-virial","Expectation Values, the Virial Theorem, and Scaling","The radial matrix elements ⟨r^k⟩ of hydrogenic states are the raw material of every later correction. This lesson derives ⟨1\u002Fr⟩ from the virial theorem, builds the full family ⟨r⟩, ⟨r²⟩, ⟨1\u002Fr²⟩, ⟨1\u002Fr³⟩ from Kramers' recursion and the Feynman-Hellmann theorem, and reads off their scaling with n, ℓ, and Z. The virial balance ⟨T⟩ = −½⟨V⟩ = −E fixes the energy budget of every bound state.\n",{"path":16338,"title":16339,"module":16319,"summary":16340},"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Fquantum-defects-alkali-spectra","Quantum Defects and Alkali Spectra","An alkali atom is one valence electron outside a closed-shell core, and to a good approximation it is hydrogen with a modified quantum number. Core penetration makes low-ℓ states more bound than the Coulomb formula predicts, and the shortfall is captured by a single number per ℓ, the quantum defect δℓ. The spectrum then follows the Rydberg formula with n replaced by the effective n − δℓ, and the sodium D-line doublet is the worked case.\n",{"path":16342,"title":16343,"module":16319,"summary":16344},"\u002Fatomic-physics\u002Fquantum-hydrogen-atom\u002Frydberg-atoms","Rydberg Atoms","A Rydberg atom is an atom excited to a very high principal quantum number, and every hydrogenic property becomes exaggerated by a power of n. Size grows as n², binding falls as n⁻², radiative lifetime lengthens as n³, and the static polarizability explodes as n⁷. The levels crowd toward the ionization limit, and the enormous dipole interaction between two Rydberg atoms produces the blockade that underlies neutral-atom quantum computing.\n",{"path":16346,"title":16347,"module":16348,"summary":16349},"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Frelativistic-kinetic-correction","The Relativistic Kinetic-Energy Correction","Fine Structure and the Dirac Atom","The Bohr energies treat the electron as slowly moving, but its speed is of order αc, so the kinetic energy needs a relativistic correction. Expanding √(p²c²+m²c⁴) to order (v\u002Fc)² produces the perturbation −p⁴\u002F8m³c², whose first-order shift on a hydrogenic state is evaluated with the trick p²=2m(E−V). The result depends on n and ℓ, is smaller than the gross structure by α²≈5×10⁻⁵, and is one of the three pieces that combine into the fine-structure formula.\n",{"path":16351,"title":16352,"module":16348,"summary":16353},"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Fspin-orbit-thomas-precession","Spin-Orbit Coupling and Thomas Precession","In the electron's rest frame the nucleus orbits it, and the resulting current produces a magnetic field that couples to the electron's spin moment. The interaction is ξ(r) L·S, with ξ built from the Coulomb potential and the radial expectation ⟨1\u002Fr³⟩. A relativistic subtlety, Thomas precession, halves the naive coefficient because the electron's rest frame is accelerating. The result splits each ℓ≥1 level into a j=ℓ±½ doublet and makes (n, ℓ, j, mⱼ) the good quantum numbers.\n",{"path":16355,"title":16356,"module":16348,"summary":16357},"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Fdarwin-term-fine-structure-formula","The Darwin Term and the Fine-Structure Formula","The third fine-structure correction, the Darwin term, is a contact interaction proportional to ∇²V that acts only on s-states, physically a smearing of the electron over a Compton wavelength. Adding the relativistic, spin-orbit, and Darwin shifts, the ℓ-dependence cancels and the total collapses to a formula in n and j alone. The n=2 shell splits into 2S₁\u002F₂, 2P₁\u002F₂, 2P₃\u002F₂, with the two j=½ levels exactly degenerate, a coincidence the Dirac theory explains.\n",{"path":16359,"title":16360,"module":16348,"summary":16361},"\u002Fatomic-physics\u002Ffine-structure-and-the-dirac-atom\u002Fdirac-equation-hydrogen","The Dirac Equation for Hydrogen","The fine-structure formula was assembled from three perturbations; the Dirac equation produces it in one stroke and exactly. A first-order relativistic wave equation forces a four-component spinor, from which spin s=½, the g-factor of 2, the spin-orbit term, and antiparticles all emerge automatically. Its exact Coulomb spectrum depends only on n and j, and expanding in Zα reproduces the perturbative result, including the 2S₁\u002F₂–2P₁\u002F₂ degeneracy that sets up the Lamb shift.\n",{"path":16363,"title":16364,"module":16365,"summary":16366},"\u002Fatomic-physics\u002Fqed-corrections-and-hyperfine-structure\u002Flamb-shift-qed","The Lamb Shift and QED Radiative Corrections","QED Corrections and Hyperfine Structure","The Dirac equation makes the 2S₁\u002F₂ and 2P₁\u002F₂ levels of hydrogen exactly degenerate. Lamb and Retherford measured a splitting of about 1058 MHz that the Dirac theory cannot produce. The gap comes from the electron's coupling to the quantized electromagnetic field: self-energy, vacuum polarization, and the anomalous magnetic moment. Welton's vacuum-fluctuation estimate reproduces the size and shows why the effect lands almost entirely on s-states, and the same radiative corrections make hydrogen the most stringent test of QED.\n",{"path":16368,"title":16369,"module":16365,"summary":16370},"\u002Fatomic-physics\u002Fqed-corrections-and-hyperfine-structure\u002Fhyperfine-structure-21cm","Hyperfine Structure and the 21 cm Line","The proton carries a magnetic moment, and it interacts with the magnetic field the electron produces at the nucleus. For s-states that interaction is the Fermi contact term, proportional to the electron density at the origin and to the dot product of the nuclear and electronic spins. Coupling I and J into F = I + J splits each level by a Landé interval rule; in hydrogen's ground state it produces the F = 0\u002FF = 1 doublet whose 1420 MHz, 21 cm transition maps neutral hydrogen across the galaxy.\n",{"path":16372,"title":16373,"module":16365,"summary":16374},"\u002Fatomic-physics\u002Fqed-corrections-and-hyperfine-structure\u002Fnuclear-effects-isotope-shift","Nuclear Size, Moments, and Isotope Shifts","A real nucleus has a finite size, a mass that changes between isotopes, and, when its spin is at least one, an electric quadrupole moment. Each leaves a fingerprint in the atomic spectrum: the volume shift from s-electrons sampling the charge distribution, the mass and field isotope shifts that separate on a King plot, the quadrupole interaction that breaks the Landé interval rule, and the hyperfine anomaly from the magnetization distribution. Atomic spectroscopy reads nuclear properties out of these shifts.\n",{"path":16376,"title":16377,"module":16378,"summary":16379},"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fperiodic-table-atomic-spectra","The Periodic Table and Atomic Spectra","Many-Electron Atoms","Identical electrons demand antisymmetric wave functions, which is the Pauli exclusion principle: no two electrons share all four quantum numbers. Filling shells in order of increasing energy — shifted by penetration and shielding — builds the periodic table and its recurring ionization pattern. Selection rules govern optical spectra, and an external field splits lines by the Zeeman effect.\n",{"path":16381,"title":16382,"module":16378,"summary":16383},"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fcentral-field-self-consistent","The Central-Field Approximation and the Self-Consistent Field","The N-electron Hamiltonian does not separate because every pair of electrons repels. The central-field approximation replaces that pairwise repulsion with an averaged spherical potential each electron feels, restoring hydrogen-like orbitals labelled by n and ℓ. The Thomas-Fermi statistical model fixes the shape of the screened charge from Fermi-gas thermodynamics; the Hartree self-consistent field determines it exactly by iterating orbitals against the potential they generate until the two agree.\n",{"path":16385,"title":16386,"module":16378,"summary":16387},"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fidentical-particles-hartree-fock","Exchange, Slater Determinants, and Hartree-Fock","A product wave function ignores that electrons are identical fermions. Enforcing antisymmetry writes the state as a Slater determinant, which vanishes whenever two electrons share a spin-orbital — the exclusion principle made algebraic. The energy of a determinant carries a new term with no classical analogue, the exchange integral, nonzero only for parallel spins; it lowers the energy of aligned electrons and carves a Fermi hole around each one. Adding the exchange operator to the mean field gives the Hartree-Fock equations, and what they still miss defines the correlation energy.\n",{"path":16389,"title":16390,"module":16378,"summary":16391},"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fhelium-two-electron-atom","Helium: the Prototype Two-Electron Atom","Helium is the smallest atom the Schrödinger equation cannot solve exactly, and the smallest that shows every many-electron effect. Ignoring the electron repulsion overbinds the ground state by 30 eV; first-order perturbation theory and a one-parameter variational calculation with an effective charge close most of the gap. The excited configurations split into para (singlet) and ortho (triplet) states separated by the exchange integral, with the triplet lower — and the absence of a 1s² triplet is the Pauli principle in its plainest form.\n",{"path":16393,"title":16394,"module":16378,"summary":16395},"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fls-jj-coupling-term-symbols","LS and jj Coupling; Term Symbols","A configuration is not a single energy level. The residual electrostatic repulsion and the spin-orbit interaction split it, and which one dominates fixes the coupling scheme. In light atoms the electrostatic term wins: orbital and spin angular momenta couple separately into L and S, then into J, giving Russell- Saunders term symbols. In heavy atoms spin-orbit wins and each electron's j forms first. The Pauli principle prunes the allowed terms of equivalent electrons, the Landé interval rule spaces the fine-structure multiplet, and the scheme crosses over from LS to jj down a column.\n",{"path":16397,"title":16398,"module":16378,"summary":16399},"\u002Fatomic-physics\u002Fmany-electron-atoms\u002Fhund-rules-ground-terms","Hund's Rules and Ground-State Terms","A configuration allows several terms; Hund's three rules pick the ground one. Maximize the spin S first, then the orbital L, then set J to |L−S| for a less-than-half shell and L+S for a more-than-half shell. The first two rules come from exchange lowering the energy of apart-kept electrons; the third comes from the sign of the spin-orbit coupling, which flips as a shell passes half-filling and turns the multiplet from normal to inverted. Worked ground terms for carbon, nitrogen, oxygen, and iron show the rules in action.\n",{"path":16401,"title":16402,"module":16403,"summary":16404},"\u002Fatomic-physics\u002Fatoms-in-external-fields\u002Fzeeman-effect","The Zeeman Effect","Atoms in External Fields","A magnetic field couples to the atom through its magnetic moment, splitting each level into equally spaced sublevels labelled by the projection of the total angular momentum. When spin is present the spacing is not the classical one: it carries the Landé g-factor, a projection of the spin and orbital moments onto the total angular momentum. We derive the weak-field Hamiltonian from minimal coupling, evaluate the shift with the projection theorem, and read off the polarization of the emitted components.\n",{"path":16406,"title":16407,"module":16403,"summary":16408},"\u002Fatomic-physics\u002Fatoms-in-external-fields\u002Fpaschen-back-intermediate","The Paschen-Back and Intermediate-Field Regimes","When the magnetic interaction grows past the fine-structure coupling, spin and orbital angular momentum decouple and precess independently about the field. The anomalous Zeeman pattern reverts to a simple triplet, the Paschen-Back effect. Between the two limits neither coupling dominates and the level positions follow from diagonalizing the combined spin-orbit and Zeeman Hamiltonian. We build the two-by-two problem for a single valence electron, solve it in closed form, and show both limits emerge from one expression.\n",{"path":16410,"title":16411,"module":16403,"summary":16412},"\u002Fatomic-physics\u002Fatoms-in-external-fields\u002Fstark-effect-polarizability","The Stark Effect and Field Ionization","An electric field shifts atomic levels by coupling to the electron's position. Parity forbids a first-order shift for a non-degenerate state, so most atoms respond only at second order through their polarizability, a quadratic Stark shift. Hydrogen is the exception: its accidental degeneracy admits a permanent dipole and a linear shift, cleanest in parabolic coordinates. At large fields the Coulomb well develops a saddle, and Rydberg states field-ionize at a threshold that falls as the fourth power of the principal quantum number.\n",{"path":16414,"title":16415,"module":16416,"summary":16417},"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Ftime-dependent-perturbation-golden-rule","Time-Dependent Perturbation Theory and the Golden Rule","Radiative Transitions and Spectral Lines","An atom in a weak oscillating field makes transitions between its stationary states. First-order time-dependent perturbation theory gives the transition amplitude as a Fourier component of the perturbation at the Bohr frequency, and the resulting probability is a sinc-squared resonance that sharpens as the field acts longer. For a two-level system the same coupling produces Rabi oscillations; for a transition into a continuum the long-time limit collapses the sinc-squared into a delta function and yields Fermi's golden rule, a constant transition rate set by the coupling strength and the density of final states.\n",{"path":16419,"title":16420,"module":16416,"summary":16421},"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Fdipole-approximation-einstein-coefficients","The Dipole Approximation and Einstein Coefficients","The coupling between an atom and light is the interaction of the electron with the electromagnetic field. Because an optical wavelength dwarfs the atom, the spatial variation of the field across the atom can be dropped, leaving the electric-dipole interaction and its matrix element. That matrix element defines the oscillator strength, which obeys the Thomas-Reiche-Kuhn sum rule. Einstein's three rate coefficients (absorption, stimulated emission, spontaneous emission) follow from detailed balance with thermal radiation, fixing the ratio of spontaneous to stimulated rates and its steep growth with frequency.\n",{"path":16423,"title":16424,"module":16416,"summary":16425},"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Fselection-rules-forbidden-transitions","Selection Rules and Forbidden Transitions","The dipole matrix element vanishes for most pairs of states, and the pattern of which survive is the set of selection rules. Parity forces the orbital angular momentum to change by one; the angular integral of three spherical harmonics restricts the magnetic quantum number to change by zero or one; the photon's spin restricts the total angular momentum. When the dipole element vanishes, higher multipoles (magnetic dipole and electric quadrupole) can still drive the transition at rates smaller by powers of the fine-structure constant, and states with no allowed decay become metastable.\n",{"path":16427,"title":16428,"module":16416,"summary":16429},"\u002Fatomic-physics\u002Fradiative-transitions-and-line-shapes\u002Flifetimes-and-line-shapes","Lifetimes, Line Widths, and Line Shapes","A spectral line is never infinitely sharp. The finite lifetime of the excited state gives every line a natural Lorentzian width set by the total decay rate, the Fourier transform of an exponentially damped emission. Thermal motion adds a Gaussian Doppler width that usually dominates in a gas; collisions add a further Lorentzian pressure width; the observed profile is the Voigt convolution of the Gaussian and Lorentzian parts. Strong driving fields broaden the line further through saturation. Each mechanism has a distinct dependence on temperature, density, and intensity that lets it be identified and, where possible, removed.\n",{"path":16431,"title":16432,"module":16433,"summary":16434},"\u002Fatomic-physics\u002Flasers-and-spectroscopy\u002Flaser-principles","Population Inversion, Gain, and the Laser","Lasers and Spectroscopy","A laser is an optical amplifier placed inside a resonant cavity. Amplification requires that stimulated emission outrun absorption, which requires more atoms in the upper level than the lower one — a population inversion that the Einstein relations forbid in thermal equilibrium and that no two-level pump can produce. Three- and four-level schemes reach it by routing atoms through auxiliary states. The gain coefficient sets how strongly a weak beam grows, the cavity fixes the threshold and selects a comb of longitudinal modes, and gain saturation clamps the steady-state inversion at its threshold value.\n",{"path":16436,"title":16437,"module":16433,"summary":16438},"\u002Fatomic-physics\u002Flasers-and-spectroscopy\u002Fspectroscopy-techniques","Spectroscopic Techniques and Frequency Combs","A tunable laser turns spectroscopy from photographing a spectrum into interrogating a single transition, but at room temperature the Doppler width buries the natural linewidth under a thousandfold-broader Gaussian. Saturated absorption and two-photon spectroscopy defeat the first-order Doppler shift by selecting the zero-velocity class or cancelling the shift between counter-propagating photons, recovering natural-width features. Laser-induced fluorescence pushes sensitivity to single atoms, and the optical frequency comb converts an optical frequency into a countable radio-frequency beat, giving absolute frequency measurement across the visible spectrum.\n",{"path":16440,"title":16441,"module":16433,"summary":16442},"\u002Fatomic-physics\u002Flasers-and-spectroscopy\u002Fline-catalog-nist-asd","Reading Real Spectra with the NIST Database","Every quantity computed in this course — energy levels, transition frequencies, oscillator strengths, lifetimes — is tabulated for real atoms in the NIST Atomic Spectra Database. This lesson reads that data as physics: how levels are labelled by term symbols and energies in wavenumbers, how a transition list encodes wavelength, Einstein coefficient, and line strength, how a Grotrian diagram is reconstructed from the tables, and how a measured spectrum is matched to catalog lines. The residual between computed and tabulated positions is the running score of atomic theory.\n",{"path":16444,"title":16445,"module":16446,"summary":16447},"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Flaser-cooling-doppler","Laser Cooling and Optical Molasses","Modern Atomic Physics","A near-resonant laser beam pushes an atom because every absorbed photon delivers one unit of momentum and the subsequent spontaneous emission averages to zero. Two counter-propagating red-detuned beams turn that push into friction: the Doppler shift brings a moving atom closer to resonance with the beam it moves against, so the net force opposes the velocity. Six beams give optical molasses in three dimensions. The random recoil of spontaneous emission heats against the friction, and the balance sets the Doppler cooling limit. Adding a magnetic-field gradient makes the force position-dependent as well, giving the magneto-optical trap.\n",{"path":16449,"title":16450,"module":16446,"summary":16451},"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Fsub-doppler-trapping","Sub-Doppler Cooling and Atom Traps","Optical molasses cools multilevel atoms below the Doppler limit. A polarization gradient plus optical pumping makes an atom repeatedly climb a light-shift hill and be pumped to the valley, losing kinetic energy each cycle — Sisyphus cooling. The floor is the recoil limit, one photon momentum of residual motion. Below it, cooling must avoid scattering photons: conservative magnetic and optical-dipole traps hold the atoms while forced evaporation removes the hot tail, driving the phase-space density up toward quantum degeneracy.\n",{"path":16453,"title":16454,"module":16446,"summary":16455},"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Fbose-einstein-condensation","Bose-Einstein Condensation of Atomic Gases","Below a critical temperature a gas of identical bosons places a macroscopic fraction of its atoms in the single lowest-energy state. The transition occurs when the thermal de Broglie wavelength grows to the interparticle spacing, so the phase- space density reaches order unity. The critical temperature follows from the Bose-Einstein distribution and the density of states, the condensate fraction grows as one minus (T\u002FTc) to the three-halves, and the condensate reveals itself in time-of-flight as a sharp bimodal peak in the momentum distribution. The 1995 rubidium and sodium experiments realized it in dilute trapped gases.\n",{"path":16457,"title":16458,"module":16446,"summary":16459},"\u002Fatomic-physics\u002Fmodern-atomic-physics\u002Foptical-clocks-precision","Optical Atomic Clocks and Precision Measurement","An atomic clock counts the oscillations of a field locked to an atomic transition. The cesium microwave standard defines the second through the 9.19 GHz ground-state hyperfine transition, interrogated by Ramsey's separated-oscillatory-field method whose fringe width is set by the free-precession time. Optical clocks replace the microwave transition with an optical one five orders of magnitude higher in frequency, raising the quality factor and the fractional stability in proportion. Lattice and single-ion clocks reach fractional uncertainties near ten-to-the-minus- eighteen by trapping the atoms at a magic wavelength that cancels the light shift, and at that level they measure the gravitational redshift over centimetres of height.\n",{"path":16461,"title":16462,"module":6,"summary":6},"\u002Fatomic-physics","Atomic Physics",{"path":16464,"title":16465,"module":6,"summary":6},"\u002Fdatabases","Databases",{"path":16467,"title":16468,"module":14054,"summary":16469},"\u002Fcategory-theory\u002Ffoundations\u002Fwhat-is-a-category","Categories, Objects, and Arrows","A category is objects, arrows between them, a rule for composing arrows, and an identity arrow on every object, subject to associativity and the unit laws. The axioms mention no elements: arrows need not be functions, and an object is known only through the arrows into and out of it. Isomorphism, commutative diagrams, duality, and the terminal object are the first consequences.\n",{"path":16471,"title":16472,"module":14054,"summary":16473},"\u002Fcategory-theory\u002Ffoundations\u002Fexamples-of-categories","A Zoo of Categories","The axioms admit two very different kinds of model: large categories of structured sets and their structure-preserving maps (Set, Mon, Grp, Top, Vect), and small categories that are themselves single algebraic objects — a monoid as a one-object category, a poset as a thin category. The awkward cases Rel and Pfn have sets as objects but relations and partial functions as arrows, and a typed programming language presents its types and programs as a category.\n",{"path":16475,"title":16476,"module":14054,"summary":16477},"\u002Fcategory-theory\u002Ffoundations\u002Fspecial-morphisms","Isomorphisms, Monos, and Epis","Injectivity and surjectivity mention elements, so a general category re-expresses them by cancellation: monomorphisms cancel on the left, epimorphisms on the right. Sections and retractions are the split versions with an explicit one-sided inverse. Mono plus epi does not force an isomorphism, and subobjects are equivalence classes of monos into a fixed object.\n",{"path":16479,"title":16480,"module":14054,"summary":16481},"\u002Fcategory-theory\u002Ffoundations\u002Ffunctors","Functors: Maps Between Categories","A functor sends objects to objects and arrows to arrows while preserving composition and identities. Covariant and contravariant functors, the standard stock (forgetful, free, hom, and powerset), and the classification by faithfulness, fullness, and essential surjectivity all follow. Functors compose, so categories and functors form a category themselves.\n",{"path":16483,"title":16484,"module":14054,"summary":16485},"\u002Fcategory-theory\u002Ffoundations\u002Fnatural-transformations","Natural Transformations and Functor Categories","A natural transformation is a map between two parallel functors: one component arrow per object, subject to a commuting square for every arrow of the source. Naturality is verified for the determinant, the double dual, and list operations; functors and natural transformations form the functor category [C, D]; and vertical and horizontal composition satisfy the Godement interchange law.\n",{"path":16487,"title":16488,"module":14054,"summary":16489},"\u002Fcategory-theory\u002Ffoundations\u002Fsize-and-set-theory","Size: Small, Large, Locally Small","The objects of Set do not form a set, and pretending otherwise reproduces the classical paradoxes. Classes make the small\u002Flarge distinction precise, with locally small and essentially small as the intermediate notions. Cantor's theorem shows Set and its algebraic relatives are large, and the function-based axiomatization of sets is the one category theory prefers to ZFC.\n",{"path":16491,"title":16492,"module":16493,"summary":16494},"\u002Fcategory-theory\u002Funiversal-properties\u002Funiversal-properties","Universal Properties, Initial and Terminal Objects","Universal Properties and Basic Constructions","A universal property characterizes an object by a for-all\u002Fexists-unique condition on the arrows into or out of it, and any two objects satisfying the same property are isomorphic by a unique isomorphism. Initial and terminal objects are the simplest cases; the free vector space, the discrete topology, and the ring of integers show the pattern at work.\n",{"path":16496,"title":16497,"module":16493,"summary":16498},"\u002Fcategory-theory\u002Funiversal-properties\u002Fproducts-and-coproducts","Products and Coproducts","The product of two objects is a wedge of projections through which every other wedge factors uniquely; the coproduct is the dual, built from injections. In Set these are the cartesian product and the disjoint union, in a poset the meet and join, and in abelian groups the two coincide. The mediating-arrow discipline established here is the template for all limits.\n",{"path":16500,"title":16501,"module":16493,"summary":16502},"\u002Fcategory-theory\u002Funiversal-properties\u002Fconstructions-on-categories","Opposite, Product, Slice, and Comma Categories","Categories are themselves mathematical structures, and the standard algebraic constructions apply: opposites, products, subcategories, slices, and the comma category that subsumes them. The opposite category yields the duality principle, halving the subject's proofs; slice and comma categories repackage every universal property as an initial or terminal object.\n",{"path":16504,"title":16505,"module":16506,"summary":16507},"\u002Fcategory-theory\u002Frepresentables-yoneda\u002Frepresentable-functors","Hom-Functors and Representables","Representables and the Yoneda Lemma","Fixing an object A of a locally small category produces a set-valued functor, the hom-functor A(A,-), that records every map out of A. A functor is representable when it is naturally isomorphic to such a hom-functor. We define the covariant and contravariant hom-functors, collect the standard representables (identity, forgetful, powerset), and read maps as generalized elements of varying shape.\n",{"path":16509,"title":16510,"module":16506,"summary":16511},"\u002Fcategory-theory\u002Frepresentables-yoneda\u002Fyoneda-lemma","The Yoneda Lemma","The Yoneda lemma computes the natural transformations out of a representable presheaf: they form a set in natural bijection with X(A). The proof fixes a single degree of freedom, the image of the identity arrow, and shows naturality forces everything else. We prove the bijection, verify naturality in both variables, and read off that a natural transformation out of a representable is just one element.\n",{"path":16513,"title":16514,"module":16506,"summary":16515},"\u002Fcategory-theory\u002Frepresentables-yoneda\u002Fyoneda-consequences","The Yoneda Embedding and Its Uses","Three corollaries turn the Yoneda lemma into working machinery. A representation of a presheaf is the same thing as a universal element; the Yoneda embedding of a category into its presheaf category is full and faithful; and two objects are isomorphic exactly when their representables are. Together they justify constructing arrows by constructing natural transformations between hom-functors, and they contain Cayley's theorem as the one-object case.\n",{"path":16517,"title":16518,"module":16519,"summary":16520},"\u002Fcategory-theory\u002Flimits-colimits\u002Flimits","Cones and Limits","Limits and Colimits","A diagram is a functor from a small shape category; a cone over it is an object with compatible legs to every node; and a limit is the terminal cone, the one every other cone factors through uniquely. Products and terminal objects reappear as limits over particular shapes, and the whole construction is unique up to a single isomorphism.\n",{"path":16522,"title":16523,"module":16519,"summary":16524},"\u002Fcategory-theory\u002Flimits-colimits\u002Fproducts-equalizers-pullbacks","Equalizers and Pullbacks","The equalizer of a parallel pair is the universal arrow that makes the two composites agree; the pullback of a cospan is the universal commutative square. In Set they are solution sets and fibered products, every equalizer is monic, monics are stable under pullback, and products plus equalizers together generate all limits.\n",{"path":16526,"title":16527,"module":16519,"summary":16528},"\u002Fcategory-theory\u002Flimits-colimits\u002Fcolimits","Colimits: Coproducts, Coequalizers, Pushouts","Colimits are limits in the opposite category: cocones replace cones, and the universal cocone is initial rather than terminal. Coproducts glue objects side by side, coequalizers impose relations and produce quotients, pushouts glue along a shared part, and in Set every colimit is a quotient of a disjoint union. Directed colimits admit a clean elementwise description.\n",{"path":16530,"title":16531,"module":16519,"summary":16532},"\u002Fcategory-theory\u002Flimits-colimits\u002Fcomputing-limits","Computing Limits in Concrete Categories","In Set the limit of any diagram is the set of threads: choice functions through the nodes that commute with every edge. In Pos, Mon, and Top the recipe is the same limit downstairs plus the unique structure that makes the projections structure-preserving — pointwise order, componentwise operations, the topology generated by the projections. The pattern is what \"the forgetful functor creates limits\" means concretely.\n",{"path":16534,"title":16535,"module":16519,"summary":16536},"\u002Fcategory-theory\u002Flimits-colimits\u002Flimits-and-functors","Preservation, Reflection, and Creation of Limits","A functor preserves limits if it sends limit cones to limit cones, reflects them if it recognizes them, and creates them if limits downstairs lift uniquely upstairs. Representable functors preserve all limits, forgetful functors from algebra create them, and limits in functor categories are computed pointwise, one evaluation at a time.\n",{"path":16538,"title":16539,"module":16540,"summary":16541},"\u002Fcategory-theory\u002Fadjunctions\u002Fadjunctions","Adjoint Functors via Hom-Set Bijections","Adjunctions","An adjunction is a natural bijection between two hom-sets: maps out of $F(A)$ in one category correspond to maps into $G(B)$ in the other. We give the definition, spell out the naturality axioms that make the correspondence compatible with composition, and work the flagship examples — free vector spaces, free groups, discrete and indiscrete topologies, and currying.\n",{"path":16543,"title":16544,"module":16540,"summary":16545},"\u002Fcategory-theory\u002Fadjunctions\u002Funits-and-counits","Units, Counits, and the Triangle Identities","The whole hom-set bijection of an adjunction is generated by two natural transformations: the unit, obtained by transposing identity maps on one side, and the counit, by transposing them on the other. Two triangle identities are all they must satisfy, and any pair satisfying them determines a unique adjunction. The same correspondence specializes to order-preserving maps between posets and to free constructions.\n",{"path":16547,"title":16548,"module":16540,"summary":16549},"\u002Fcategory-theory\u002Fadjunctions\u002Fadjunctions-via-universal-arrows","Adjunctions from Universal Arrows","The unit component at a single object is an initial object of a comma category, and this universal property alone rebuilds the whole adjunction. A functor has a left adjoint exactly when every object admits such a universal arrow, and the left adjoint is assembled from them one object at a time. We prove the equivalence of all three formulations of adjointness.\n",{"path":16551,"title":16552,"module":16540,"summary":16553},"\u002Fcategory-theory\u002Fadjunctions\u002Ffree-forgetful-adjunctions","Free Constructions and Free–Forgetful Adjunctions","Free monoids, free groups, and free vector spaces are left adjoints to forgetful functors, and the universal mapping property is all one needs to prove it. Some forgetful functors also have right adjoints (co-free constructions like the indiscrete topology), producing three-functor chains. Contravariant adjunctions, symmetric in their two functors, close the lesson with the pattern behind duality and representation theorems.\n",{"path":16555,"title":16556,"module":16557,"summary":16558},"\u002Fcategory-theory\u002Fadjoints-limits\u002Flimits-via-adjoints","Limits as Adjoints and as Representables","Adjoints, Representables, and Limits Together","A cone on a diagram is a natural transformation from a constant diagram, so a limit is a representation of the cone functor and, equivalently, a value of the right adjoint to the diagonal functor. We prove both rephrasings, derive uniqueness and functoriality of limits from them, and record the dual statement that a colimit is the left adjoint to the diagonal.\n",{"path":16560,"title":16561,"module":16557,"summary":16562},"\u002Fcategory-theory\u002Fadjoints-limits\u002Fpresheaf-limits-colimits","Limits and Colimits of Presheaves","Representables preserve limits, and limits in a functor category are computed one object at a time, so a presheaf category is complete and cocomplete with all its structure inherited pointwise from Set. The Yoneda embedding then preserves limits but not colimits, and the density theorem repairs the colimit side: every presheaf is a canonical colimit of representables.\n",{"path":16564,"title":16565,"module":16557,"summary":16566},"\u002Fcategory-theory\u002Fadjoints-limits\u002Fadjoints-preserve-limits","Right Adjoints Preserve Limits (RAPL)","A functor with a left adjoint preserves every limit that exists, and dually a functor with a right adjoint preserves colimits. The proof is a four-line chain of natural isomorphisms through the adjunction and the continuity of representables. The theorem yields product-and-exponential arithmetic in Set, another proof that limits commute with limits, and a standard test for proving that a functor has no adjoint.\n",{"path":16568,"title":16569,"module":16557,"summary":16570},"\u002Fcategory-theory\u002Fadjoints-limits\u002Fadjoint-functor-theorem","The Adjoint Functor Theorem","RAPL makes limit preservation necessary for having a left adjoint; the adjoint functor theorems identify when it is sufficient. For ordered sets no extra hypothesis is needed. In general the candidate adjoint is a limit over a comma category that may be large, and the general adjoint functor theorem tames it with a weakly initial set. We prove GAFT in full and apply it to free groups and, through the special adjoint functor theorem, the Stone–Čech compactification.\n",{"path":16572,"title":16573,"module":16574,"summary":16575},"\u002Fcategory-theory\u002Fmonads-algebras\u002Fmonads","Monads from Adjunctions","Monads and Algebras","A monad on a category is an endofunctor equipped with a unit and a multiplication satisfying associativity and unit laws — the data of a monoid, written internally to the category of endofunctors. Every adjunction induces one, and the list, exception, and state constructions that model computational effects are all monads on Set.\n",{"path":16577,"title":16578,"module":16574,"summary":16579},"\u002Fcategory-theory\u002Fmonads-algebras\u002Falgebras-eilenberg-moore","Algebras for a Monad","An algebra for a monad is an object with a structure map that interacts correctly with the unit and multiplication. The algebras form the Eilenberg–Moore category, whose free–forgetful adjunction induces the monad back; a comparison functor relates any other inducing adjunction to it, and for the list monad the algebras are exactly monoids.\n",{"path":16581,"title":16582,"module":16574,"summary":16583},"\u002Fcategory-theory\u002Fmonads-algebras\u002Fkleisli-and-programming","The Kleisli Category and Monads in Programming","The Kleisli category of a monad has the same objects as the base but takes arrows A to TB, composed by mapping and flattening. These arrows are effectful programs, Kleisli composition is the bind of functional programming, and the Kleisli adjunction is the initial resolution of the monad, with Eilenberg–Moore at the terminal end.\n",{"path":16585,"title":16586,"module":16574,"summary":16587},"\u002Fcategory-theory\u002Fmonads-algebras\u002Falgebras-for-endofunctors","Algebras for an Endofunctor and Recursion","Dropping the monad laws leaves algebras for a bare endofunctor, whose initial objects are the least fixed points of the functor by Lambek's lemma. The natural numbers, lists, and trees are initial algebras; the unique map out of an initial algebra is the fold of functional programming; and the Smyth–Plotkin fixed-point technique builds Scott domains the same way.\n",{"path":16589,"title":16590,"module":16591,"summary":16592},"\u002Fcategory-theory\u002Fcartesian-closed-lambda\u002Fcartesian-closed-categories","Cartesian Closed Categories","Cartesian Closed Categories and Typed Lambda Calculus","A cartesian closed category has a terminal object, binary products, and for every pair of objects an exponential object that internalizes the hom-set as an object of the category. The defining data is an evaluation arrow and a currying operation, packaged by the adjunction between product-with-A and exponential-by-A. Set, Boolean and Heyting algebras, functor categories, and Cat are all cartesian closed.\n",{"path":16594,"title":16595,"module":16591,"summary":16596},"\u002Fcategory-theory\u002Fcartesian-closed-lambda\u002Flambda-calculus-correspondence","Typed Lambda Calculus and CCCs","The typed lambda calculus and the cartesian closed category are two presentations of the same theory. Types become objects, terms with one free variable become arrows, product types become products, and function types become exponentials, with abstraction matching currying and application matching evaluation. Building the category of a lambda theory and the internal language of a category are mutually inverse up to equivalence.\n",{"path":16598,"title":16599,"module":16591,"summary":16600},"\u002Fcategory-theory\u002Fcartesian-closed-lambda\u002Ffixed-points-and-recursion","Fixed Points in Cartesian Closed Categories","The untyped lambda calculus has a fixed-point combinator; the typed calculus cannot, and Lawvere's fixed-point theorem explains why: any point-surjection onto an exponential forces every endomap to have a fixed point, which is the abstract form of Cantor's diagonal argument. Recursion is recovered instead by restricting to omega-complete partially ordered objects, where every continuous endomap has a least fixed point built by iterating from bottom. This gives While loops a semantics.\n",{"path":16602,"title":16603,"module":6,"summary":6},"\u002Fcategory-theory","Category Theory",{"path":16605,"title":15247,"module":16606,"summary":16607},"\u002Fdeep-learning\u002Fmathematical-background\u002Flinear-algebra-for-deep-learning","Mathematical Background","Every quantity a network touches is a tensor, and every layer is a matrix acting on one. This lesson compiles the linear algebra deep learning actually uses: products and norms, the system $Ax=b$ and when it is solvable, the two decompositions (eigen and SVD) that diagonalize a transformation, and the pseudoinverse that solves what cannot be solved exactly. It then derives PCA as the worked example that ties it all together.\n",{"path":16609,"title":16610,"module":16606,"summary":16611},"\u002Fdeep-learning\u002Fmathematical-background\u002Fprobability-and-information-theory","Probability & Information Theory","This lesson assembles the probabilistic vocabulary a network is trained in (random variables, densities, the chain rule, expectation and covariance, the handful of distributions that recur everywhere) and then the information theory that turns a probabilistic model into a loss: self-information, entropy, and the KL divergence whose asymmetry is the cross-entropy objective itself.\n",{"path":16613,"title":16614,"module":16606,"summary":16615},"\u002Fdeep-learning\u002Fmathematical-background\u002Fnumerical-computation","Numerical Computation","Machine learning runs on finite-precision arithmetic, where every number is approximated and every operation rounds. This lesson sets the numerical ground rules: overflow and underflow and the standard stabilizations, the condition number that measures how much a problem amplifies error, and the gradient-based optimization (first and second order, constrained and unconstrained) that every training loop runs.\n",{"path":16617,"title":14606,"module":16606,"summary":16618},"\u002Fdeep-learning\u002Fmathematical-background\u002Fcalculus","This lesson assembles the differential calculus used in training networks: the gradient and directional derivative, the Jacobian and Hessian, and the chain rule in scalar, vector, and matrix form. From the chain rule it derives back-propagation as a single sweep over the computational graph, tabulates the matrix-calculus identities that recur in layer gradients, reads optimization off a second-order Taylor expansion, and ends with why reverse-mode automatic differentiation is the algorithm every framework runs.\n",{"path":16620,"title":16621,"module":14054,"summary":16622},"\u002Fdeep-learning\u002Ffoundations\u002Fwhat-is-deep-learning","What Is Deep Learning?","Deep learning is representation learning by composition: stack simple differentiable layers, define a loss, and let gradient descent discover the features a human would otherwise have to engineer by hand. We set up the whole vocabulary (model, loss, optimizer, data), the training loop that ties them together, and the three reasons the approach became practical.\n",{"path":16624,"title":16625,"module":14054,"summary":16626},"\u002Fdeep-learning\u002Ffoundations\u002Fmachine-learning-refresher","A Machine-Learning Refresher","The statistical framework the networks live in: data drawn from an unknown distribution, a loss to minimize, and the central question of generalization: will it work on data we have not seen? We set up empirical risk, capacity, the bias–variance tradeoff, and maximum likelihood.\n",{"path":16628,"title":16629,"module":14054,"summary":16630},"\u002Fdeep-learning\u002Ffoundations\u002Flinear-models-and-the-perceptron","Linear Models & the Perceptron","The simplest learners (linear regression, logistic regression, the perceptron) already contain the whole template: a weighted sum, a loss, a gradient step. They also fail on the XOR problem, which no linear model can solve — the limitation that motivates deep learning.\n",{"path":16632,"title":16633,"module":16634,"summary":16635},"\u002Fdeep-learning\u002Fneural-networks\u002Fthe-multilayer-perceptron","The Multilayer Perceptron","Neural Networks","Stacking linear layers with a nonlinearity between them removes the limitation that stopped the perceptron. We build the multilayer perceptron in explicit matrix form (the forward pass, its dimensions, a worked XOR network with concrete weights) and prove why the nonlinearity is essential: without it the deepest stack collapses to a single hyperplane.\n",{"path":16637,"title":16638,"module":16634,"summary":16639},"\u002Fdeep-learning\u002Fneural-networks\u002Factivation-functions","Activation Functions","The activation is the only nonlinear part of a layer, and the reason depth adds expressive power. We catalog the standard hidden units (sigmoid, tanh, ReLU and its descendants, plus GELU, softplus, swish and maxout), derive each unit's derivative in full, make the vanishing-gradient problem quantitative with the chain-rule product, work numeric examples, and explain why the saturating units gave way to ReLU and why ReLU's own dead-unit failure gave way to Leaky\u002FPReLU\u002FELU\u002FGELU.\n",{"path":16641,"title":16642,"module":16634,"summary":16643},"\u002Fdeep-learning\u002Fneural-networks\u002Funiversal-approximation","Universal Approximation","One hidden layer with a non-polynomial activation can approximate any continuous function on a compact set to arbitrary accuracy: the universal approximation theorem. We prove it constructively (two sigmoids make a bump; sums of bumps make any curve), then show the limitation: existence is not efficiency. Depth-separation results exhibit functions a deep net represents with $O(n)$ units that a shallow net needs $\\exp(n)$ units to match.\n",{"path":16645,"title":16646,"module":16634,"summary":16647},"\u002Fdeep-learning\u002Fneural-networks\u002Fbackpropagation","Backpropagation","Backpropagation is the chain rule run backward over a computational graph. We formalize the graph, derive the four backprop equations for an MLP, present the forward and backward passes as algorithms, and work a tiny two-layer net by hand with explicit numbers. The result: one scalar loss, reverse-mode autodiff, and a gradient for every parameter at twice the cost of a forward pass.\n",{"path":16649,"title":16650,"module":16634,"summary":16651},"\u002Fdeep-learning\u002Fneural-networks\u002Floss-functions-and-output-units","Loss Functions & Output Units","The last layer is where a network's hidden representation meets the task. Choosing an output unit and a loss is not two independent choices; maximum likelihood fixes the pair. We derive the standard couplings (linear\u002FMSE, sigmoid\u002FBCE, softmax\u002Fcross-entropy), show why softmax and cross-entropy were built to cancel into the residual $\\hat y - y$, and prove why squared error is the wrong loss for a saturating classifier.\n",{"path":16653,"title":16654,"module":16655,"summary":16656},"\u002Fdeep-learning\u002Foptimization\u002Fgradient-descent-and-sgd","Gradient Descent & SGD","Optimization","Training is descent on the empirical risk: step the parameters against the gradient. We derive the minibatch gradient as an unbiased estimator whose variance falls as $1\u002FB$, derive the learning-rate ceiling from the smoothness-stability bound $\\eta \u003C 2\u002FL$, and lay out the schedules (step, exponential, cosine, warmup) that anneal it over training.\n",{"path":16658,"title":16659,"module":16655,"summary":16660},"\u002Fdeep-learning\u002Foptimization\u002Fmomentum-and-adaptive-methods","Momentum & Adaptive Methods","Plain gradient descent zig-zags across ravines and moves slowly along flat valleys, because one global learning rate cannot suit a surface with wildly different curvature in different directions. Two fixes address the two problems: momentum accumulates a velocity that damps the oscillation and accelerates the drift, and adaptive methods give every parameter its own learning rate scaled by the history of its gradients. Adam fuses both, and is the default optimizer of modern deep learning.\n",{"path":16662,"title":16663,"module":16655,"summary":16664},"\u002Fdeep-learning\u002Foptimization\u002Finitialization","Weight Initialization","The initial weights determine whether training can succeed before the first gradient step. Initialize every weight equal and all hidden units compute the same function forever; initialize too small or too large and the signal vanishes or explodes as it crosses depth. A single variance condition, $n_{\\text{in}}\\mathrm{Var}(W)=1$, fixes both, and reading it off the forward and backward passes yields Xavier and He initialization directly.\n",{"path":16666,"title":16667,"module":16655,"summary":16668},"\u002Fdeep-learning\u002Foptimization\u002Fthe-optimization-landscape","The Optimization Landscape","The loss of a deep network is a non-convex surface in millions of dimensions, so local search carries no global guarantee, yet it works. We classify critical points by the eigenvalues of the Hessian, show that in high dimension nearly all of them are saddle points rather than bad local minima, and read off the practical terrain — plateaus, cliffs, ill-conditioning, and the sharp-versus-flat distinction that ties the geometry of a minimum to how well it generalizes.\n",{"path":16670,"title":16671,"module":16655,"summary":16672},"\u002Fdeep-learning\u002Foptimization\u002Fsecond-order-and-approximate-methods","Second-Order & Approximate Methods","Newton's method reads the curvature of the loss off its Hessian and jumps to the minimum of the local quadratic in a single step, rescaling away the ill-conditioning that slows first-order descent. We derive it, then explain the three obstacles that keep it out of deep learning: a $d \\times d$ Hessian for $d$ in the billions, an attraction to saddle points, and minibatch noise. The alternative is approximation (conjugate gradients, BFGS and L-BFGS, the natural gradient and Hessian-free methods), each buying some of Newton's curvature information without ever forming or inverting $H$.\n",{"path":16674,"title":16675,"module":16676,"summary":16677},"\u002Fdeep-learning\u002Fregularization\u002Fregularization-overview","Regularization Overview","Regularization","Regularization is any modification to a learning algorithm meant to lower test error at the possible expense of training error. We derive the bias–variance decomposition that explains why it helps, set up the two parameter-norm penalties, $L^2$ weight decay and $L^1$, derive their update rules and eigenbasis shrinkage, show geometrically why $L^1$ alone produces sparse weights (soft-thresholding), distinguish weight decay from loss-added $L^2$ under AdamW, and read both penalties through the two lenses that recur across the chapter: a norm-ball constraint via KKT, and a prior via MAP estimation.\n",{"path":16679,"title":16680,"module":16676,"summary":16681},"\u002Fdeep-learning\u002Fregularization\u002Fdropout-and-data-augmentation","Dropout & Data Augmentation","Two of the most effective regularizers add no penalty term at all; they perturb the computation instead. Dropout multiplies hidden units by a random Bernoulli mask, training an exponential ensemble of thinned subnetworks that share weights; inverted scaling collapses that ensemble into one cheap forward pass at test time. Data augmentation enlarges the training set with label-preserving transforms, injecting the invariances the task demands, and noise injection (input, weight, label smoothing, Mixup) generalizes the same idea into a continuous family.\n",{"path":16683,"title":16684,"module":16676,"summary":16685},"\u002Fdeep-learning\u002Fregularization\u002Fearly-stopping-and-parameter-sharing","Early Stopping & Parameter Sharing","Two cheap regularizers that cost no extra term in the loss. Early stopping treats training time itself as a hyperparameter (watch the validation curve, halt at its minimum, keep the best checkpoint), and for a quadratic objective it is provably equivalent to $L^2$ weight decay. Parameter sharing goes the other way: it constrains many weights to be _equal_, the prior behind every convolution and every recurrent step, and the reason a CNN has orders of magnitude fewer parameters than the dense net it replaces.\n",{"path":16687,"title":16688,"module":16676,"summary":16689},"\u002Fdeep-learning\u002Fregularization\u002Fnormalization","Normalization","Normalization layers standardize activations to zero mean and unit variance inside the network, then hand the model a learnable scale and shift to undo the constraint when it pays to. Batch normalization does this across the batch and must keep separate train-time and test-time statistics; layer, instance, and group norm change only the axes they average over. The result is faster, better-conditioned optimization and a free dose of regularizing batch noise.\n",{"path":16691,"title":16692,"module":16693,"summary":16694},"\u002Fdeep-learning\u002Farchitectures\u002Fconvolutional-networks","Convolutional Networks","Architectures","A convolutional network replaces the dense layer's all-to-all weight matrix with a small kernel slid across the input. Three structural commitments (sparse connectivity, parameter sharing, and translation equivariance) collapse the parameter count by orders of magnitude and bake the right prior for images directly into the architecture. We derive the convolution arithmetic, the output geometry, pooling, and the receptive field, then assemble the canonical stack.\n",{"path":16696,"title":16697,"module":16693,"summary":16698},"\u002Fdeep-learning\u002Farchitectures\u002Fcnn-architectures","CNN Architectures","Six landmark networks, each contributing exactly one idea: LeNet's conv-pool stack, AlexNet's ReLU-and-dropout scale, VGG's $3\\times3$ uniformity, Inception's multi-scale module, ResNet's residual skip, and DenseNet's dense connectivity. The common thread is the degradation problem (why plain deeper nets train worse, not just overfit) and the residual block that solved it by keeping a $+1$ path open for the gradient.\n",{"path":11729,"title":16700,"module":16693,"summary":16701},"Recurrent Networks","A recurrent network folds a sequence into a fixed-size hidden state, reusing one set of weights at every time step, the architectural prior that the same rule applies wherever it lands in time. Unrolling the recurrence exposes a deep feed-forward graph; backpropagation through it sums gradient contributions across all steps and chains a product of Jacobians, and that product is why long-range gradients vanish or explode. That failure motivates gated architectures.\n",{"path":11724,"title":11725,"module":16693,"summary":16703},"A plain recurrent network propagates its hidden state through a repeated weight-matrix multiply, and the Jacobian product that results vanishes or explodes long before a useful gradient can reach the early steps. Gated RNNs fix this with an additive memory path: a cell state that is carried forward almost unchanged, past which the gradient flows along a near-identity highway. We derive that highway, give the full LSTM and GRU equations, and compare the two.\n",{"path":16705,"title":16706,"module":16693,"summary":16707},"\u002Fdeep-learning\u002Farchitectures\u002Fattention-and-transformers","Attention & Transformers","Attention replaces fixed wiring with content-based routing: every position reads from every other through a soft, learned dot-product lookup. We derive scaled dot-product attention and its $\\sqrt{d_k}$ correction, build it into multi-head self-attention, inject order with positional encodings, and stack the whole thing into the Transformer block that displaced recurrence and convolution alike.\n",{"path":16709,"title":16710,"module":16693,"summary":16711},"\u002Fdeep-learning\u002Farchitectures\u002Fthe-transformer-architecture","The Transformer Architecture","The Transformer is the architecture built around the attention mechanism. This first part assembles the full encoder–decoder of \"Attention Is All You Need\" — embeddings and positional encoding, stacked self-attention and feed-forward sublayers wrapped in residual connections and LayerNorm, masked decoding and cross-attention — works through causal masking and the three modern families (encoder-only, decoder-only, encoder–decoder), and accounts for where the parameters and the $O(n^2)$ compute actually go.\n",{"path":16713,"title":16714,"module":16693,"summary":16715},"\u002Fdeep-learning\u002Farchitectures\u002Ftransformers-in-practice","Transformers in Practice","The Transformer makes no assumption about what a token represents. This part follows the architecture out of language: image patches feed a plain encoder (the Vision Transformer), the decoder-only half scales into the GPT line of large language models, and one substrate covers translation, retrieval, and multimodal grounding. We work the ViT patch arithmetic and a GPT parameter count by hand, then close on the empirical scaling laws — power-law loss, the Chinchilla compute-optimal balance, and emergent behavior — that made scale the dominant lever.\n",{"path":16717,"title":16718,"module":16693,"summary":16719},"\u002Fdeep-learning\u002Farchitectures\u002Fgraph-neural-networks","Graph Neural Networks","A graph neural network learns on data with no grid and no canonical ordering: atoms in a molecule, users in a social network, road segments in a map. The unifying idea is message passing — each node repeatedly aggregates its neighbors' states and updates its own — built to respect the one symmetry graphs demand, permutation equivariance. We derive the message-passing framework, specialize it into GCN, GraphSAGE, GAT, and GIN, read off graph-level outputs, and bound what message passing can and cannot tell apart.\n",{"path":16721,"title":16722,"module":16693,"summary":16723},"\u002Fdeep-learning\u002Farchitectures\u002Fstate-space-models","State-Space Models and Mamba","A state-space model carries a continuous linear hidden state through a sequence, and that linearity buys two equivalent algorithms from one set of weights: a recurrence that runs in linear time with constant memory, and a global convolution that trains in parallel. Long-range memory comes from how the transition matrix is initialized (HiPPO) and parameterized (S4's diagonal-plus-low-rank form). Mamba breaks the convolution on purpose, making the parameters input-dependent so the model can select what to remember, recovered at speed by a hardware-aware parallel scan.\n",{"path":16725,"title":16726,"module":16727,"summary":16728},"\u002Fdeep-learning\u002Ftheory\u002Fgeneralization-theory","Generalization Theory","Theory & Frontiers","Classical learning theory bounds the gap between training and test error by a model's capacity (VC dimension, Rademacher complexity), and predicts that a model with more parameters than data should overfit catastrophically. Modern networks do the opposite: they interpolate, even fit pure noise, and still generalize. We derive the classical bounds, work the bias-variance decomposition, show why the bounds go vacuous, and survey what replaced them: double descent, the interpolation threshold, margin and norm-based bounds, and the implicit bias of the optimizer itself.\n",{"path":16730,"title":16731,"module":16727,"summary":16732},"\u002Fdeep-learning\u002Ftheory\u002Fadversarial-robustness","Adversarial Robustness","A trained network can be fooled by a perturbation too small for a human to see: add a carefully aimed vector of magnitude $\\epsilon$ to a correctly classified image and the prediction flips. We derive the fast gradient sign method as the first-order-optimal step inside an $L_\\infty$ ball, explain the linearity hypothesis that makes high-dimensional models so easy to push around, build up to projected gradient descent, and frame adversarial training as a min-max robust-optimization problem with its own accuracy cost. Defenses beyond training continue in the next lesson.\n",{"path":16734,"title":16735,"module":16727,"summary":16736},"\u002Fdeep-learning\u002Ftheory\u002Fadversarial-defenses","Adversarial Defenses","Defending a network against an adversary is far harder than attacking one. This lesson covers the defense side: certified guarantees via randomized smoothing, the transferability that makes black-box attacks possible, and the recurring failure of gradient masking, where a defense hides the attacker's gradient instead of moving the decision boundary. It ends with the adaptive-attack discipline (BPDA, EOT, transfer) that every robustness claim must be tested against.\n",{"path":16738,"title":16739,"module":16727,"summary":16740},"\u002Fdeep-learning\u002Ftheory\u002Fbayesian-and-ensemble-methods","Bayesian & Ensemble Methods","A trained network returns a single point prediction and, with the softmax, a confidence, but that confidence is usually miscalibrated, collapsing to near- certainty even on inputs the model has never seen. This lesson covers uncertainty estimation for networks: the two kinds of uncertainty, the Bayesian posterior over weights and its tractable stand-ins (MC dropout, deep ensembles), and how to check whether a model's reported confidences match observed frequencies.\n",{"path":16742,"title":16743,"module":16727,"summary":16744},"\u002Fdeep-learning\u002Ftheory\u002Fdeep-equilibrium-models","Deep Equilibrium Models","A deep network need not be a fixed stack of layers; it can be a single weight-tied layer iterated to convergence, its output defined implicitly as the fixed point $z^\\star = f_\\theta(z^\\star, x)$. The forward pass becomes root-finding and the backward pass becomes implicit differentiation, so training costs O(1) memory regardless of effective depth. We derive both passes from the implicit function theorem and close the course on defining a layer by a fixed-point condition rather than an explicit stack.\n",{"path":16746,"title":16747,"module":16748,"summary":16749},"\u002Fdeep-learning\u002Fgenerative-models\u002Flinear-factor-models","Linear Factor Models","Generative Models","The simplest generative models share one template: a latent variable drawn from a fixed prior, run through a linear decoder, plus noise. Probabilistic PCA, factor analysis, independent component analysis, and sparse coding are all this template with a different prior on the latents and a different noise model. We derive each marginal, see why ICA needs non-Gaussianity to identify its sources, and show how sparse coding learns Gabor-like dictionary atoms.\n",{"path":16751,"title":16752,"module":16748,"summary":16753},"\u002Fdeep-learning\u002Fgenerative-models\u002Fautoencoders","Autoencoders","An autoencoder is a network trained to copy its input to its output through a narrow channel; the useful product is the bottleneck representation $h$, not the reconstruction. We derive the undercomplete autoencoder and prove its linear case recovers PCA, then trade the bottleneck for explicit regularization (sparse, denoising, contractive) and show how a denoising autoencoder learns the low-dimensional manifold the data lives on.\n",{"path":16755,"title":16756,"module":16748,"summary":16757},"\u002Fdeep-learning\u002Fgenerative-models\u002Fvariational-autoencoders","Variational Autoencoders","An autoencoder compresses, but its latent space has gaps: sample a point between two encodings and the decoder produces noise. The variational autoencoder fixes this by training a probabilistic encoder against a prior, so the latent space becomes a smooth, samplable density. We derive the evidence lower bound it maximizes, the reparameterization trick that lets gradients flow through a random sample, and the closed-form Gaussian regularizer that pulls the posterior toward the prior.\n",{"path":16759,"title":16760,"module":16748,"summary":16761},"\u002Fdeep-learning\u002Fgenerative-models\u002Fgenerative-adversarial-networks","Generative Adversarial Networks","A generative adversarial network trains two networks against each other: a generator that turns noise into samples, and a discriminator that tries to tell real data from forgeries. The game has a clean theory: the optimal discriminator is a likelihood ratio, and at equilibrium the generator minimizes the Jensen–Shannon divergence to the data, with a global optimum exactly when its distribution matches the data. We derive that result, fix the saturating loss that breaks training, and catalogue the failure modes (mode collapse, instability, vanishing gradients) and the architectural fixes.\n",{"path":16763,"title":16764,"module":16748,"summary":16765},"\u002Fdeep-learning\u002Fgenerative-models\u002Fautoregressive-and-normalizing-flows","Autoregressive Models & Normalizing Flows","Two families that provide exact likelihoods, each at a cost. Autoregressive models factor the joint by the probability chain rule and learn each conditional with a masked network: exact $\\log p(x)$, but sampling proceeds one coordinate at a time. Normalizing flows push a simple base density through an invertible map and read $\\log p(x)$ off the change-of-variables formula, trading architectural freedom for a cheap Jacobian determinant via triangular coupling layers.\n",{"path":16767,"title":16768,"module":16748,"summary":16769},"\u002Fdeep-learning\u002Fgenerative-models\u002Fenergy-based-and-boltzmann-machines","Energy-Based & Boltzmann Machines","Energy-based models replace an explicit density with a scalar energy and a Boltzmann normalization, $p(x) = e^{-E(x)}\u002FZ$: simple to specify, but with an intractable partition function $Z$. The Boltzmann machine and its restricted variant make the energy bilinear so the hidden units factorize, and contrastive divergence sidesteps $Z$ by replacing the model expectation with a few Gibbs steps started at the data. We close on the undirected deep models (DBNs and DBMs) and how they differ from the directed VAE.\n",{"path":16771,"title":16772,"module":16748,"summary":16773},"\u002Fdeep-learning\u002Fgenerative-models\u002Fdiffusion-and-score-based-models","Diffusion and Score-Based Models","Corrupt a data point with Gaussian noise in small steps until only noise remains, then train a network to undo one step at a time. We derive the forward process and its closed-form marginal, reduce the variational bound to the single noise-prediction objective that makes diffusion trainable, and show the score-matching view that unifies it with Langevin sampling and the continuous SDE. The lesson closes with DDIM fast sampling, classifier-free guidance, and the latent diffusion that powers modern text-to-image systems.\n",{"path":16775,"title":16776,"module":16777,"summary":16778},"\u002Fdeep-learning\u002Fprobabilistic-methods\u002Fstructured-probabilistic-models","Structured Probabilistic Models","Probabilistic Methods","A joint distribution over $n$ variables is a table with exponentially many entries; nobody can store it, fit it, or sample from it directly. Structure fixes this: a graph whose missing edges encode conditional independencies that factor the joint into small local pieces. We build the two dialects, directed (Bayesian networks) and undirected (Markov random fields), read independence off the graph, and connect the machinery to the latent-variable and energy-based models that power deep generative learning.\n",{"path":16780,"title":16781,"module":16777,"summary":16782},"\u002Fdeep-learning\u002Fprobabilistic-methods\u002Fmonte-carlo-and-mcmc","Monte Carlo & MCMC","Most quantities of interest in a probabilistic model are integrals nobody can compute in closed form: expectations, marginals, partition functions. Monte Carlo replaces the integral with an average over samples; importance sampling reweights samples from a tractable proposal; and when even sampling the target is hard, Markov-chain Monte Carlo builds a chain whose stationary distribution _is_ the target. We derive Metropolis–Hastings and Gibbs, analyze mixing, and close on the partition-function gradient that powers energy-based learning.\n",{"path":16784,"title":16785,"module":16777,"summary":16786},"\u002Fdeep-learning\u002Fprobabilistic-methods\u002Fapproximate-inference","Approximate Inference","In a latent-variable model the quantity we need, the posterior $p(h\\mid v)$ over hidden causes, is almost never computable, because its normalizer is an intractable sum over configurations. Approximate inference reframes the problem as optimization: maximize the evidence lower bound, a tractable functional whose gap to the true log-evidence equals a KL divergence. From that single bound fall expectation–maximization, mean-field variational inference, MAP, and the learned encoders behind variational autoencoders.\n",{"path":16788,"title":16789,"module":16790,"summary":16791},"\u002Fdeep-learning\u002Fpractical\u002Fpractical-methodology","Practical Methodology","Practical Deep Learning","Knowing the algorithms is half the job; the other half is a disciplined loop. Fix a goal and a metric, stand up an end-to-end baseline, then read the train\u002Fvalidation gap to decide whether the next move is more data or a bigger model. We detail that loop: choosing metrics under class imbalance, default baselines by data type, extrapolating the data a target needs, and guarding the data pipeline against the leaks and label bugs that corrupt every gradient. Hyperparameter tuning, debugging, and deployment continue in the sequel.\n",{"path":16793,"title":16794,"module":16790,"summary":16795},"\u002Fdeep-learning\u002Fpractical\u002Fhyperparameters-and-debugging","Hyperparameters & Debugging","The tuning half of the methodology loop. The learning rate is the one hyperparameter that dominates, so we tune it first, on a log scale, coarse to fine, and prefer random search to grid when only a few dials matter. Then an ordered debugging playbook — overfit one batch, check the loss at initialization against ln C, watch the gradient norm, gradient-check against centered finite differences — and, after launch, monitoring for train-test skew and distribution drift with confidence-based abstention.\n",{"path":16797,"title":16798,"module":16790,"summary":16799},"\u002Fdeep-learning\u002Fpractical\u002Frepresentation-learning","Representation Learning","A good representation makes a hard task easy by changing coordinates: it disentangles the factors of variation, spends its bits as a distributed code, and respects the low-dimensional manifold the data lives on. We make those three properties precise, recover the manifold hypothesis, and close on the first method that turned them into training practice — greedy layer-wise unsupervised pretraining — before the sequel picks up how the field learned to reuse those features.\n",{"path":16801,"title":16802,"module":16790,"summary":16803},"\u002Fdeep-learning\u002Fpractical\u002Ftransfer-learning","Transfer Learning","A representation learned once can be reused everywhere. We cover the main mechanisms of reuse: feature extraction versus fine-tuning, the generic-to-specific gradient of features that sets the freeze boundary, the learning-rate discipline that keeps borrowed weights from being erased, domain adaptation when only the input distribution shifts, and the modern arc from supervised transfer to self-supervised foundation models.\n",{"path":16805,"title":16806,"module":16790,"summary":16807},"\u002Fdeep-learning\u002Fpractical\u002Fapplications","Applications","We survey large-scale training (the hardware, the two axes of parallelism, mixed precision, and the compression tricks that shrink a model after it is trained), then specialize the same gradient loop to vision, language, speech, and recommendation. Each domain is a different prior bolted onto one optimizer: convolutional invariance for pixels, distributed word vectors for tokens, sequence transduction for audio, low-rank factorization for the user–item matrix.\n",{"path":16809,"title":16810,"module":16790,"summary":16811},"\u002Fdeep-learning\u002Fpractical\u002Fmodel-compression-and-distillation","Model Compression and Distillation","A trained network and a deployable one are rarely the same object. This lesson is the toolkit for closing that gap: knowledge distillation transfers a large teacher's soft, information-rich logits into a small student; pruning deletes the weights that contribute least; quantization swaps 32-bit floats for 8- or 4-bit integers; and low-rank factorization replaces a fat matrix with two thin ones. We derive each method, show what it costs in accuracy, and lay out which combinations win on which hardware.\n",{"path":16813,"title":16814,"module":16790,"summary":16815},"\u002Fdeep-learning\u002Fpractical\u002Fmeta-learning-and-few-shot","Meta-Learning and Few-Shot Learning","A deep network trained on one example per class overfits. Meta-learning targets this few-shot regime by training across a distribution of tasks so that a new task is learnable from a handful of examples. We formalize the $N$-way $K$-shot episode, then derive the two dominant families: metric methods that learn an embedding where distance classifies (Prototypical Networks), and optimization methods that learn an initialization a few gradient steps can adapt (MAML). We close on the link to transfer learning and to the in-context few-shot behavior of large language models.\n",{"path":16817,"title":16818,"module":16819,"summary":16820},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Flarge-language-models","Large Language Models","Large Models & Agents","A large language model is a decoder-only Transformer trained on one objective, next-token prediction, then scaled until new behavior appears. This first part builds the object itself: the equivalence between next-token prediction and lossless compression, subword tokenization (BPE, WordPiece, Unigram, SentencePiece) worked on a real sentence, the four pretraining objectives and the attention masks that distinguish them, and the three model families (encoder-only, decoder-only, encoder--decoder) with their parameter budgets. Scaling, decoding, the KV cache, and alignment continue in part two.\n",{"path":16822,"title":16823,"module":16819,"summary":16824},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fscaling-inference-and-alignment","Scaling, Inference, and Alignment of Language Models","Once a language model is built, three questions remain: how does it improve as it grows, how is it decoded and served affordably, and how is a raw next-token predictor turned into an assistant. We derive the Kaplan power laws and the Chinchilla compute-optimal balance, trace emergent abilities and in-context learning, catalog the decoding strategies from greedy to nucleus sampling, work the KV cache that makes generation quadratic instead of cubic, cover parameter-efficient adaptation by low-rank updates (LoRA), and close on the alignment stack: instruction tuning, RLHF, and DPO.\n",{"path":16826,"title":16827,"module":16819,"summary":16828},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fseq2seq-pretraining-and-bart","Denoising Sequence-to-Sequence Pretraining: BART","BERT corrupts and reconstructs; GPT predicts the next token. Sequence-to-sequence pretraining unifies both by training a full encoder–decoder as a denoising autoencoder: corrupt the text with a noise function, then reconstruct the original through a bidirectional encoder and an autoregressive decoder. This first part derives the denoising objective, catalogs BART's five noise functions (with a worked Poisson-infilling budget), proves BART specializes to both BERT and GPT, and traces a dimension-annotated forward pass through its encoder--decoder. T5, PEGASUS, fine-tuning, and decoding continue in part two.\n",{"path":16830,"title":16831,"module":16819,"summary":16832},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Ftext-to-text-transfer-and-conditional-generation","Text-to-Text Transfer and Conditional Generation","BART reconstructs a corrupted document; T5 pushes the same denoising idea into a single interface where every task is a string-to-string map. This second part covers T5's span corruption with sentinel tokens (with a worked token budget), PEGASUS's summarization-matched gap sentences and the MASS midpoint, supervised fine-tuning and beam-search decoding with a length penalty, the exposure-bias failure modes of autoregressive decoding, and a theorem showing why a bidirectional encoder--decoder strictly dominates a decoder-only model when the output is conditioned on a full input.\n",{"path":16834,"title":16835,"module":16819,"summary":16836},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fspeech-and-audio-models","Speech Recognition: Front-Ends and Alignment","Speech is a long, high-rate sequence whose label is short and unaligned, so the whole subject turns on bridging that mismatch. This first part builds the spectral front-ends that compress a waveform into frames (STFT, mel spectrogram, MFCC, with a worked frame-count), derives CTC's marginalization over alignments and its forward-backward recursion with a two-frame numeric example, and contrasts it with attention-based seq2seq (LAS) and the RNN transducer. Self-supervised and weakly-supervised models, and text-to-speech, continue in part two.\n",{"path":16838,"title":16839,"module":16819,"summary":16840},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fself-supervised-speech-and-synthesis","Self-Supervised Speech Models and Synthesis","The recognition front-ends and alignment losses of part one all need transcribed audio, which is scarce. This second part removes that dependence: wav2vec 2.0 learns speech representations from unlabeled audio by a masked contrastive objective, HuBERT swaps the contrast for masked prediction of clustered units, and Whisper trades curation for scale with weakly-supervised web audio and a multitask token interface. We close with text-to-speech (the same length mismatch run backwards) and a tour of speech foundation models, discrete audio codecs, and neural TTS.\n",{"path":16842,"title":16843,"module":16819,"summary":16844},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fai-agents","AI Agents: Tools and Reasoning","A language model that only emits text is a function from prompt to prompt; an agent closes the loop, letting that model act on an environment, read back the result, and decide again. This first part formalizes the agent as a policy over interaction histories, builds out tool calling and the executor trust boundary, the ReAct interleaving of reasoning and action (with concrete traces), and search over thoughts: chain-of-thought, self-consistency, least-to-most, and Tree of Thoughts. Memory, retrieval, reflection, and multi-agent orchestration continue in part two.\n",{"path":16846,"title":16847,"module":16819,"summary":16848},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fagent-memory-retrieval-and-orchestration","Agent Memory, Retrieval, and Orchestration","An agent's reasoning and tool use only matter if it can remember what it learned and coordinate work larger than one context window. This second part builds the systems around the loop: short-term scratchpad versus long-term vector store, retrieval-augmented generation with a worked softmax over passage scores, reflection (Reflexion, Self-Refine), and multi-agent orchestration. It closes on the failure modes that bound agents — invalid tool calls, horizon-error compounding, context overflow, non-terminating loops — and the benchmarks that score the full loop.\n",{"path":16850,"title":16851,"module":16819,"summary":16852},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fmixture-of-experts","Mixture-of-Experts","A mixture-of-experts layer replaces one feed-forward network with many and a router that sends each token to only a few of them, so the parameter count and the per-token compute become separate dials. We derive the gated output, sparse top-$k$ routing softmax, the load-balancing loss that stops the router from collapsing onto a single expert, and expert\u002Ftoken capacity with dropping, then work the dimension-annotated tensor shapes and FLOP arithmetic. We trace the architectures from the sparsely-gated LSTM through GShard, Switch Transformer, and Mixtral, cover distributed expert parallelism, and close on the training dynamics, failure modes, and serving costs of a sparse model.\n",{"path":16854,"title":16855,"module":16819,"summary":16856},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Fmultimodal-models","Multimodal Contrastive Learning","A multimodal model places images, text, and audio in one representation space, so a picture and its caption land close together. This first part builds the contrastive route: the shared embedding space and its residual modality gap, the Vision Transformer image encoder (patch embedding, CLS token, position embeddings, with shapes), the symmetric InfoNCE loss that trains the CLIP dual encoder from a batch similarity matrix (with a worked numeric step), and zero-shot classification as a softmax over class-prompt embeddings. Fusion and vision-language models continue in part two.\n",{"path":16858,"title":16859,"module":16819,"summary":16860},"\u002Fdeep-learning\u002Flarge-models-and-agents\u002Ffusion-and-vision-language-models","Fusion and Vision-Language Models","A contrastive model compares modalities but never lets one read another. This second part builds the fusion route: early, late, and cross-attention fusion, then the three designs that connect a frozen vision encoder to a frozen language model — Flamingo's zero-initialized gated cross-attention, BLIP-2's Q-Former, and LLaVA's linear projector. We work the token-budget arithmetic that separates them, name the object-hallucination and fine-detail failure modes, cover the contrastive-then- instruction-tune recipe and its retrieval\u002Fcaptioning\u002FVQA benchmarks, and close on natively multimodal models.\n",{"path":16862,"title":16863,"module":16864,"summary":16865},"\u002Fdeep-learning\u002Freinforcement-learning\u002Ffoundations-of-reinforcement-learning","Foundations of Reinforcement Learning","Reinforcement Learning","Reinforcement learning is the third paradigm: an agent learns to act by interacting with an environment that returns rewards, not labels. We formalize the interaction as a Markov decision process, define the value functions that rank states and actions, and derive the Bellman expectation and optimality equations that every method downstream solves. Dynamic programming gives the exact answer when the model is known, and its convergence rests on a single fact: the Bellman operator is a contraction.\n",{"path":16867,"title":16868,"module":16864,"summary":16869},"\u002Fdeep-learning\u002Freinforcement-learning\u002Fmodel-free-prediction-and-control","Model-Free Prediction and Control","When the dynamics are unknown, an agent cannot plan against a model; it must learn directly from sampled experience. We build prediction and control from two estimators of the same return: Monte Carlo averages whole episodes, while temporal-difference learning bootstraps from its own next estimate. We trace the bias-variance contrast between them, derive SARSA and Q-learning as the on-policy and off-policy forms of control, unify everything through n-step returns and eligibility traces, and close on the deadly triad that makes off-policy bootstrapping with function approximation diverge.\n",{"path":16871,"title":16872,"module":16864,"summary":16873},"\u002Fdeep-learning\u002Freinforcement-learning\u002Fdeep-q-networks","Deep Q-Networks","A Deep Q-Network replaces the tabular action-value function with a neural approximator $Q(s,a;\\theta)$ and trains it by regression toward a bootstrapped target. Naive online Q-learning with a network diverges, so DQN adds two stabilizers: an experience-replay buffer that decorrelates samples, and a periodically-frozen target network that holds the regression target still. We derive the loss, give the full algorithm and the Atari pipeline, and then layer on Double DQN, the dueling split, prioritized replay, and the Rainbow combination.\n",{"path":16875,"title":16876,"module":16864,"summary":16877},"\u002Fdeep-learning\u002Freinforcement-learning\u002Fpolicy-gradients-and-actor-critic","Policy Gradients and Actor-Critic Methods","Value-based reinforcement learning learns what each state is worth and acts greedily; policy-gradient methods skip the detour and optimize a parameterized policy directly by ascending the gradient of expected return. The policy gradient theorem makes this tractable through the log-derivative trick, turning an intractable gradient of an expectation into an expectation of a gradient. REINFORCE realizes the idea but suffers high variance; baselines, the advantage function, and actor-critic learning reduce it, and trust-region methods (TRPO, PPO) keep each update from destroying the policy it just learned.\n",{"path":16879,"title":16880,"module":16864,"summary":16881},"\u002Fdeep-learning\u002Freinforcement-learning\u002Frl-from-human-feedback","Reinforcement Learning from Human Feedback","Many objectives we want from a model, that it be helpful and harmless, are hard to write down but easy to judge by comparison. RLHF turns that asymmetry into a training signal: fit a reward model to pairwise human preferences under the Bradley-Terry likelihood, then fine-tune the policy to maximize that reward under a KL penalty toward a reference. We derive the reward loss, the KL-regularized RL objective and its closed-form optimum, then show how DPO inverts that optimum to collapse the whole pipeline into one supervised log-sigmoid loss, and survey IPO, KTO, RLAIF, and GRPO.\n",{"path":16883,"title":16884,"module":6,"summary":6},"\u002Fdeep-learning","Deep Learning",{"path":16886,"title":16887,"module":14797,"summary":16888},"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fequilibrium-state-variables-zeroth-law","Equilibrium, State Variables, and the Zeroth Law","Thermodynamics describes a many-body system by a handful of macroscopic variables and the equilibrium relations among them. This lesson fixes the vocabulary: systems and the walls that separate them, state variables versus path-dependent process quantities, quasi-static and reversible idealizations, and the zeroth law, whose transitivity of thermal equilibrium is what lets temperature exist as a number. The ideal-gas thermometer turns that number into a scale, and an equation of state ties the variables into a surface.\n",{"path":16890,"title":16891,"module":14797,"summary":16892},"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Ffirst-law-heat-and-work","The First Law: Internal Energy, Heat, and Work","The first law is energy conservation for a system that exchanges energy as both heat and work. Internal energy is a state function with an exact differential; heat and work are path-dependent process quantities. This lesson states $\\d U=\\delta Q+\\delta W$, computes compression work as an area on the $P$–$V$ plane, defines the heat capacities $C_V$ and $C_P$ and the enthalpy that makes $C_P$ natural, and works the isothermal and adiabatic processes of an ideal gas, including the adiabat $PV^\\gamma=\\text{const}$.\n",{"path":16894,"title":16895,"module":14797,"summary":16896},"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fsecond-law-entropy-and-the-carnot-bound","The Second Law, Carnot Cycles, and Entropy","The second law forbids the free conversion of heat into work. This lesson states the Kelvin and Clausius forms, proves them equivalent, and analyzes the Carnot cycle to get the efficiency bound $1-T_c\u002FT_h$. Carnot's theorem makes that bound universal and defines the thermodynamic temperature scale. The Clausius inequality $\\oint \\delta Q\u002FT\\le 0$ then constructs entropy as a state function, $\\d S=\\delta Q_{\\rm rev}\u002FT$, whose non-decrease in isolated systems is the arrow of time.\n",{"path":16898,"title":16899,"module":14797,"summary":16900},"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fthermodynamic-potentials-and-maxwell-relations","Thermodynamic Potentials and Maxwell Relations","The fundamental relation $\\d U=T\\,\\d S-P\\,\\d V+\\mu\\,\\d N$ packages the first and second laws into one exact differential. Legendre transforms swap each conjugate pair to produce the Helmholtz, enthalpy, Gibbs, and grand potentials, each minimized under its own natural variables. Equality of mixed second partials of these potentials gives the Maxwell relations, which convert unmeasurable entropy derivatives into measurable ones from the equation of state.\n",{"path":16902,"title":16903,"module":14797,"summary":16904},"\u002Fstatistical-mechanics\u002Fthermodynamics\u002Fstability-response-functions-and-the-third-law","Response Functions, Stability, and the Third Law","Response functions — heat capacities, compressibilities, thermal expansion — are the second derivatives of the potentials and the quantities an experiment actually measures. This lesson derives the general relation $C_P-C_V=TV\\alpha^2\u002F\\kappa_T$, shows that convexity of the potentials forces the stability conditions $C_V>0$ and $\\kappa_T>0$, and states the third law: entropy approaches a constant as $T\\to0$, so heat capacities and expansion coefficients vanish there and absolute zero is unattainable.\n",{"path":16906,"title":16907,"module":16908,"summary":16909},"\u002Fstatistical-mechanics\u002Ffoundations\u002Fclassical-statistics-and-equipartition","Classical Statistics and Equipartition","Microstates, Phase Space, and Statistical Entropy","A liter of gas holds on the order of a trillion trillion molecules, far too many to track by their equations of motion. Classical statistical mechanics replaces the trajectories with a single probability law, the Boltzmann distribution, and reads the measurable properties of matter off it: the Maxwell speed distribution, the average energy per degree of freedom, and the heat capacities of gases and solids — together with the low-temperature failures that forced the quantum revision.\n",{"path":16911,"title":16912,"module":16908,"summary":16913},"\u002Fstatistical-mechanics\u002Ffoundations\u002Fphase-space-and-liouvilles-theorem","Phase Space, Trajectories, and Liouville's Theorem","A classical system of N particles is one point in a 6N-dimensional phase space, and its evolution is a single trajectory driven by Hamilton's equations. This lesson builds that geometric picture, introduces the phase-space density of an ensemble, and proves Liouville's theorem: the density is carried by the flow as an incompressible fluid, so phase-space volume is conserved. The stationary densities of equilibrium follow as functions of the conserved quantities alone.\n",{"path":16915,"title":16916,"module":16908,"summary":16917},"\u002Fstatistical-mechanics\u002Ffoundations\u002Fensembles-and-the-equal-probability-postulate","Ensembles and the Postulate of Equal a Priori Probabilities","An ensemble is a probability distribution over the microstates of a system. This lesson states the single postulate on which equilibrium statistical mechanics rests — that an isolated system in equilibrium is equally likely to be in any of its accessible microstates — and works out its consequences: the accessible phase-space volume, the overwhelming dominance of the most probable macrostate as the particle number grows, and the ergodic hypothesis that lets a time average be replaced by an ensemble average.\n",{"path":16919,"title":16920,"module":16908,"summary":16921},"\u002Fstatistical-mechanics\u002Ffoundations\u002Fstatistical-entropy-boltzmann-and-gibbs","Statistical Entropy: Boltzmann and Gibbs","Entropy is the logarithm of the number of accessible microstates. This lesson builds the two statistical entropies — Boltzmann's S = k ln Omega for an isolated system and Gibbs's S = -k sum p ln p for any ensemble — proves they agree for a uniform distribution, and connects both to Shannon's measure of missing information. The second law emerges as the drift toward maximum multiplicity, and maximizing the Gibbs entropy under constraints previews the canonical distribution.\n",{"path":16923,"title":16924,"module":16925,"summary":16926},"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Fmicrocanonical-ensemble-and-entropy","The Microcanonical Ensemble and Statistical Entropy","The Microcanonical Ensemble","An isolated system holds its energy, volume, and particle number fixed, and the fundamental postulate assigns equal probability to every microstate on its energy shell. This lesson builds the microcanonical distribution, defines the enclosed phase-space volume $\\Gamma(E)$, the surface density of states $\\omega(E)=\\d\\Gamma\u002F\\d E$, and the shell count $\\Omega(E)$, shows their logarithms agree to $O(\\ln N)$ for large $N$, and reads the Boltzmann entropy $S=k\\ln\\Omega$ off the count. The measure factors $h^{3N}$ and $N!$ enter here and make $S$ extensive.\n",{"path":16928,"title":16929,"module":16925,"summary":16930},"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Fequilibrium-conditions-temperature-pressure-chemical-potential","Thermal, Mechanical, and Diffusive Equilibrium","Two isolated subsystems that can exchange energy, volume, or particles reach equilibrium at the partition that maximizes their combined entropy. Setting the derivative of the total entropy to zero identifies the statistical definitions $1\u002FT=(\\partial S\u002F\\partial E)$, $P\u002FT=(\\partial S\u002F\\partial V)$, and $-\\mu\u002FT=(\\partial S\u002F\\partial N)$, shows heat flows from hot to cold as an entropy increase, and recovers the fundamental relation $\\d S=(\\d E+P\\,\\d V-\\mu\\,\\d N)\u002FT$ from pure counting.\n",{"path":16932,"title":16933,"module":16925,"summary":16934},"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Fideal-gas-phase-space-and-the-sackur-tetrode-entropy","The Ideal Gas, Phase-Space Volume, and the Sackur–Tetrode Entropy","The monatomic ideal gas is the first system whose microcanonical count can be done in closed form. The momentum integral is the volume of a $3N$-dimensional ball of radius $\\sqrt{2mE}$, the configuration integral is $V^N$, and together they give the Sackur–Tetrode entropy $S=Nk[\\ln(V\u002FN\\lambda^3)+5\u002F2]$ with the thermal wavelength $\\lambda=h\u002F\\sqrt{2\\pi mkT}$. The formula matches the measured entropy of helium, fixes the classical regime $n\\ll n_Q$, and shows why the $N!$ is needed for extensivity.\n",{"path":16936,"title":16937,"module":16925,"summary":16938},"\u002Fstatistical-mechanics\u002Fmicrocanonical\u002Ftwo-state-systems-paramagnets-and-negative-temperature","Two-State Systems, Paramagnets, and Negative Temperature","The ideal two-state paramagnet has a multiplicity counted by the binomial coefficient, an entropy that is an inverted dome in the energy, and a temperature read from the slope $1\u002FT=\\partial S\u002F\\partial E$. Because the energy is bounded above, the slope changes sign past the entropy maximum: a population-inverted spin system has a negative absolute temperature, which is hotter than any positive temperature. Nuclear-spin experiments and lasers realize the inverted state.\n",{"path":16940,"title":16941,"module":16942,"summary":16943},"\u002Fstatistical-mechanics\u002Fcanonical\u002Fcanonical-ensemble-and-the-boltzmann-distribution","The Canonical Ensemble and the Boltzmann Distribution","The Canonical Ensemble","A system held at fixed temperature by contact with a heat reservoir is described by the canonical ensemble. Expanding the reservoir entropy to first order in the system energy gives the Boltzmann distribution $p_i\\propto e^{-\\beta E_i}$, and the same law follows from maximizing the Gibbs entropy at fixed mean energy. Both routes identify $\\beta=1\u002Fk_BT$ and fix the probability of every microstate from the temperature alone.\n",{"path":16945,"title":16946,"module":16942,"summary":16947},"\u002Fstatistical-mechanics\u002Fcanonical\u002Fpartition-function-and-the-helmholtz-free-energy","The Partition Function and the Helmholtz Free Energy","The normalizing sum of the Boltzmann distribution, the partition function $Z=\\sum_i e^{-\\beta E_i}$, is a generating function for the thermodynamics. The mean energy is $-\\partial\\ln Z\u002F\\partial\\beta$, and the Gibbs entropy of the canonical distribution collapses to the bridge relation $F=-k_BT\\ln Z$. From $F$ every thermodynamic quantity follows by differentiation, and $Z$ factorizes over independent degrees of freedom.\n",{"path":16949,"title":16950,"module":16942,"summary":16951},"\u002Fstatistical-mechanics\u002Fcanonical\u002Fenergy-fluctuations-and-ensemble-equivalence","Energy Fluctuations and the Equivalence of Ensembles","In the canonical ensemble the energy fluctuates, and the second derivative of $\\ln Z$ gives its variance. The fluctuation–response identity $\\langle\\Delta E^2\\rangle = k_BT^2C_V$ ties the spread of the energy to the heat capacity, and the relative fluctuation falls as $1\u002F\\sqrt{N}$. In the thermodynamic limit the canonical energy distribution is a sharp spike, and the canonical and microcanonical ensembles predict the same thermodynamics.\n",{"path":16953,"title":16954,"module":16942,"summary":16955},"\u002Fstatistical-mechanics\u002Fcanonical\u002Fthe-einstein-solid-and-harmonic-systems","Harmonic Systems: The Einstein Solid and Vibrational Heat Capacity","A quantum harmonic oscillator has a geometric partition function summed in closed form, giving a mean energy $\\hbar\\omega(\\tfrac12+\\langle n\\rangle)$ with the Bose occupation factor. Modeling a solid as $3N$ independent oscillators yields a heat capacity that rises from zero and saturates at the Dulong–Petit value $3Nk_B$. The Einstein temperature sets the crossover, and the model's exponential low-temperature falloff, too steep against the observed $T^3$, motivates the Debye theory.\n",{"path":16957,"title":16958,"module":16942,"summary":16959},"\u002Fstatistical-mechanics\u002Fcanonical\u002Fparamagnetism-and-the-schottky-anomaly","Paramagnetism, Two-Level Systems, and the Schottky Anomaly","A magnetic moment in a field is a two-level system whose partition function is a hyperbolic cosine. The magnetization of a spin-$\\tfrac12$ paramagnet is $N\\mu\\tanh(\\mu B\u002Fk_BT)$, generalizing to the Brillouin function for spin $J$; it gives Curie's law $\\chi\\propto 1\u002FT$ at high temperature and saturates at low temperature. A finite level gap produces the Schottky heat-capacity peak, and the temperature dependence of the entropy on the field is the basis of adiabatic demagnetization cooling.\n",{"path":16961,"title":16962,"module":16963,"summary":16964},"\u002Fstatistical-mechanics\u002Fclassical-gas\u002Fideal-gas-partition-function-and-the-gibbs-paradox","The Ideal Gas Partition Function and the Gibbs Paradox","The Classical Ideal Gas","The classical monatomic ideal gas built from the partition function. The single-particle sum is $z_1=V\u002F\\lambda^3$ with the thermal de Broglie wavelength $\\lambda$; the $N$-particle partition function is $z_1^N\u002FN!$, and the $N!$ is forced by indistinguishability. From $Z$ the ideal-gas law, $U=\\tfrac32 Nk_BT$, and the Sackur–Tetrode entropy follow. The $N!$ makes the entropy extensive and resolves the Gibbs paradox: mixing identical gases produces no entropy change.\n",{"path":16966,"title":16967,"module":16963,"summary":16968},"\u002Fstatistical-mechanics\u002Fclassical-gas\u002Fequipartition-and-the-virial-theorem","Equipartition and the Virial Theorem","The equipartition theorem derived from the canonical ensemble: every phase-space coordinate that enters the Hamiltonian quadratically carries a mean energy $\\tfrac12 k_BT$. The generalized form $\\langle x_i\\,\\partial H\u002F\\partial x_j\\rangle = k_BT\\,\\delta_{ij}$ contains equipartition and the classical virial theorem as special cases. Equipartition fixes the classical heat capacities, fails by quantum freeze-out when a level gap exceeds $k_BT$, and shifts for a relativistic gas whose energy is linear rather than quadratic in momentum.\n",{"path":16970,"title":16971,"module":16963,"summary":16972},"\u002Fstatistical-mechanics\u002Fclassical-gas\u002Fmolecular-gases-rotation-and-vibration","Molecular Gases: Rotational and Vibrational Degrees of Freedom","The internal partition function of a diatomic gas factorizes into translational, rotational, vibrational, and electronic parts. The rigid rotor gives a rotational temperature $\\theta_{\\rm rot}$; the harmonic bond gives a vibrational temperature $\\theta_{\\rm vib}$. Each mode contributes to the heat capacity only above its characteristic temperature, producing the diatomic $C_V$ staircase from $\\tfrac32 R$ to $\\tfrac52 R$ to $\\tfrac72 R$. Homonuclear molecules carry a symmetry number, and hydrogen splits into ortho and para species.\n",{"path":16974,"title":16975,"module":16976,"summary":16977},"\u002Fstatistical-mechanics\u002Fgrand-canonical\u002Fgrand-canonical-ensemble-and-the-grand-partition-function","The Grand Canonical Ensemble","Grand Canonical Ensemble","When a system exchanges both energy and particles with a reservoir, the reservoir fixes its temperature and its chemical potential. Expanding the reservoir entropy to first order in the exchanged energy and particle number gives the Gibbs factor $e^{-\\beta(E-\\mu N)}$, and summing it over every microstate of every particle number gives the grand partition function $\\Xi$. The grand potential $\\Phi = -k_BT\\ln\\Xi = -PV$ generates the mean particle number, energy, entropy, and pressure by differentiation.\n",{"path":16979,"title":16980,"module":16976,"summary":16981},"\u002Fstatistical-mechanics\u002Fgrand-canonical\u002Fchemical-potential-fugacity-and-number-fluctuations","Chemical Potential, Fugacity, and Number Fluctuations","The chemical potential is the energy to add one particle at fixed entropy and volume, equal to the slope of the free energy in the particle number. For the classical ideal gas $\\mu=k_BT\\ln(n\\lambda^3)$ is large and negative, and the fugacity $z=n\\lambda^3$ is small. The grand ensemble makes the particle number fluctuate; its variance $\\langle\\Delta N^2\\rangle=k_BT(\\partial N\u002F\\partial\\mu)$ equals $k_BT\\,N^2\\kappa_T\u002FV$, tying density fluctuations to the isothermal compressibility. Equality of $\\mu$ is the condition for diffusive equilibrium and phase coexistence.\n",{"path":16983,"title":16984,"module":16976,"summary":16985},"\u002Fstatistical-mechanics\u002Fgrand-canonical\u002Fensemble-summary-and-the-thermodynamic-web","The Three Ensembles and the Thermodynamic Web","The microcanonical, canonical, and grand canonical ensembles hold different variables fixed and generate different potentials — the entropy $S$, the Helmholtz free energy $F$, and the grand potential $\\Phi$ — linked by Legendre transforms that trade each fixed variable for its conjugate. Each successive ensemble lets one more quantity fluctuate. In the thermodynamic limit the three agree, the relative fluctuations vanishing as $1\u002F\\sqrt{N}$; the ideal gas gives the same equation of state in all three. The choice of ensemble is a matter of convenience, set by which sum is easiest.\n",{"path":16987,"title":16988,"module":16989,"summary":16990},"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fquantum-statistics-bose-einstein-and-fermi-dirac","Quantum Statistics — Bose-Einstein and Fermi-Dirac","Quantum Statistics","Quantum particles of the same kind are genuinely indistinguishable: no label survives an overlap of their wave functions. Counting states with that constraint replaces the Boltzmann distribution with two quantum laws — the Bose-Einstein distribution for integer-spin particles, which clump into shared states, and the Fermi-Dirac distribution for half-integer-spin particles, which exclude one another. Both reduce to Boltzmann in the dilute, hot limit, and a de Broglie criterion says exactly when.\n",{"path":16992,"title":16993,"module":16989,"summary":16994},"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fderiving-the-quantum-distributions","Deriving the Quantum Distributions from the Grand Ensemble","The Bose-Einstein and Fermi-Dirac distributions follow from one observation: in the occupation-number representation the single-particle modes are independent, so the grand partition function factorizes into one factor per mode. A boson mode sums a geometric series over all occupancies; a fermion mode sums two terms. Differentiating each factor gives the mean occupation $1\u002F(e^{\\beta(\\varepsilon-\\mu)}\\mp 1)$, the Maxwell-Boltzmann limit when occupancies are small, and the occupation fluctuations that distinguish bunching from anti-bunching.\n",{"path":16996,"title":16997,"module":16989,"summary":16998},"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fthe-classical-limit-and-quantum-concentration","The Classical Limit and Quantum Concentration","When every single-particle level is nearly empty, both quantum distributions collapse to the Maxwell-Boltzmann form, and the fugacity equals the ratio of the number density to the quantum concentration $n_Q = 1\u002F\\lambda^3$. The gas is classical when $n \\ll n_Q$, degenerate when $n \\gtrsim n_Q$. The chemical potential is large and negative in the classical regime and rises through zero as the gas degenerates. The leading quantum correction to the ideal-gas law is a second virial term that lowers the pressure for bosons and raises it for fermions — a statistical attraction and repulsion with no interaction behind it.\n",{"path":17000,"title":17001,"module":16989,"summary":17002},"\u002Fstatistical-mechanics\u002Fquantum-statistics\u002Fideal-quantum-gases-general-framework","Ideal Quantum Gases: The General Framework","Every ideal quantum gas is handled by one calculation. The sum over single-particle modes becomes an energy integral weighted by a density of states $g(\\varepsilon)\\propto\\varepsilon^{1\u002F2}$, and the number and pressure reduce to the Bose and Fermi functions $g_\\nu(z)$ and $f_\\nu(z)$ of the fugacity. An integration by parts fixes $PV=\\tfrac23 U$ for a nonrelativistic gas and $PV=\\tfrac13 U$ for an ultrarelativistic one, independent of statistics. Specializing the density of states and the chemical potential then produces the photon gas, phonons, the Bose gas, and the Fermi gas as four branches of the same framework.\n",{"path":17004,"title":17005,"module":17006,"summary":17007},"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fbose-einstein-condensation-and-the-fermion-gas","Bose-Einstein Condensation and the Fermion Gas","Bosonic Systems","Below a critical temperature a boson gas drops a macroscopic fraction of its particles into the single ground state — Bose-Einstein condensation, the mechanism behind superfluid helium and the dilute-atom condensates cooled to nanokelvin. The same statistics applied to a photon gas reproduces Planck's blackbody spectrum. Fermions do the opposite: forbidden from sharing states, they fill every level up to the Fermi energy, and that filled sea governs the electrons in metals and the pressure that holds up a white dwarf.\n",{"path":17009,"title":17010,"module":17006,"summary":17011},"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fthe-photon-gas-and-plancks-radiation-law","The Photon Gas and Planck's Radiation Law","Electromagnetic radiation in equilibrium with cavity walls is a gas of non-conserved bosons, and non-conservation forces the chemical potential to zero. Counting standing-wave modes with two polarizations and weighting each by the Bose occupation gives the Planck spectral energy density. Its low-frequency tail reproduces the classical Rayleigh-Jeans law and the ultraviolet catastrophe; the Bose factor cuts the divergence off at high frequency and the peak obeys Wien's displacement law.\n",{"path":17013,"title":17014,"module":17006,"summary":17015},"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fblackbody-thermodynamics-and-radiation-pressure","Blackbody Thermodynamics and Radiation Pressure","Integrating the Planck spectrum over all frequencies gives the total energy density proportional to the fourth power of temperature — the Stefan-Boltzmann law — and the isotropy of a relativistic gas fixes the radiation pressure at one third of the energy density. From the free energy follow the entropy and heat capacity, both proportional to T cubed, and the adiabatic law for radiation. The results govern the pressure inside stars and the cooling of the cosmic microwave background as the universe expands.\n",{"path":17017,"title":17018,"module":17006,"summary":17019},"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fphonons-and-the-debye-model","Phonons and the Debye Model","The vibrations of a crystal lattice are quantized into phonons — bosons of zero chemical potential, counted exactly like cavity photons but with three polarizations, a finite sound speed, and a total of 3N modes. The Debye model replaces the true dispersion by a linear one cut off at a frequency that enforces that count. It gives the correct low-temperature T-cubed heat capacity the Einstein model missed and recovers the Dulong-Petit value at high temperature.\n",{"path":17021,"title":17022,"module":17006,"summary":17023},"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fbose-einstein-condensation-derived","Bose-Einstein Condensation Derived","For a gas of conserved bosons the excited states can hold only a finite number of particles at fixed temperature, set by the Bose function at unit fugacity. When the total exceeds that ceiling the surplus collapses into the single ground state, which the continuum density-of-states integral misses and which must be restored by hand. This fixes the critical temperature, the condensate fraction, and the fact that a uniform gas condenses only in three or more dimensions.\n",{"path":17025,"title":17026,"module":17006,"summary":17027},"\u002Fstatistical-mechanics\u002Fbose-systems\u002Fthermodynamics-of-the-bose-gas-and-superfluidity","Thermodynamics of the Bose Gas and Superfluidity","The energy and pressure of the ideal Bose gas follow from the Bose function at the order above the density, and below the critical temperature the pressure depends on temperature alone because the condensate carries none. The heat capacity rises to a cusp at the transition. Real superfluid helium departs from the ideal gas because interactions matter: the Landau criterion ties frictionless flow to the phonon-roton excitation spectrum, and the two-fluid model carries a second sound.\n",{"path":17029,"title":17030,"module":17031,"summary":17032},"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fthe-ideal-fermi-gas-at-zero-temperature","The Ideal Fermi Gas at Zero Temperature","Degenerate Fermi Gas","At absolute zero a gas of non-interacting fermions fills every single-particle state up to the Fermi energy and leaves the rest empty, a filled Fermi sphere in momentum space. This lesson computes the Fermi momentum, energy, and temperature from the density, the density of states, the total ground-state energy, and the degeneracy pressure that grows as $n^{5\u002F3}$. Numerical Fermi energies for metals set the scale: they are electron-volts, so room temperature is deep in the degenerate regime.\n",{"path":17034,"title":17035,"module":17031,"summary":17036},"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fsommerfeld-expansion-and-electrons-in-metals","The Sommerfeld Expansion and Electrons in Metals","Turning on a small temperature blurs the Fermi step over a shell of width $k_BT$ around $\\epsilon_F$. The Sommerfeld expansion turns integrals over the Fermi function into a power series in $(k_BT\u002F\\epsilon_F)^2$, giving the shift of the chemical potential and a heat capacity linear in $T$. This resolves the old puzzle of the missing electronic heat capacity, predicts the combined $C=\\gamma T+AT^3$ of a metal, and gives the temperature-independent Pauli paramagnetism of the electron gas.\n",{"path":17038,"title":17039,"module":17031,"summary":17040},"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fwhite-dwarfs-and-the-chandrasekhar-limit","White Dwarfs and the Chandrasekhar Limit","A white dwarf is held up against its own gravity by the degeneracy pressure of its electrons. Balancing that pressure against gravity gives a mass-radius relation $R\\propto M^{-1\u002F3}$: heavier white dwarfs are smaller and denser. As the density rises the electrons turn relativistic, the pressure softens from $n^{5\u002F3}$ to $n^{4\u002F3}$, and the star can no longer support itself above a critical mass. This lesson derives that Chandrasekhar mass, about $1.4\\,M_\\odot$, and what lies beyond it.\n",{"path":17042,"title":17043,"module":17031,"summary":17044},"\u002Fstatistical-mechanics\u002Ffermi-gas\u002Fneutron-stars-and-nuclear-matter","Neutron Stars and Dense Matter","When a collapsing core passes nuclear density, electron capture converts the matter to neutrons and their degeneracy pressure takes over. The same balance that fixes a white dwarf, rescaled by the neutron mass, gives a neutron star of a few solar masses in a ten-kilometre radius. General relativity is no longer a correction: the Tolman-Oppenheimer-Volkoff equation replaces the Newtonian balance and sets a maximum mass around two solar masses. This lesson rescales the Fermi-gas argument, states where it breaks, and places the compact objects in one stability sequence.\n",{"path":17046,"title":17047,"module":17048,"summary":17049},"\u002Fstatistical-mechanics\u002Finteractions\u002Fthe-cluster-expansion-and-virial-coefficients","The Cluster Expansion and Virial Coefficients","Interacting Gases","A real gas departs from $PV=Nk_BT$ because its molecules interact. The configuration integral factors through the Mayer function $f_{ij}=e^{-\\beta u_{ij}}-1$, and expanding it in powers of density produces the virial expansion $PV\u002FNk_BT = 1 + B_2(T)n + B_3(T)n^2 + \\cdots$. The second virial coefficient $B_2(T)=-\\tfrac12\\int f\\,\\d^3r$ is a single integral over the pair potential; it is positive for a hard core, negative for an attractive well, and vanishes at the Boyle temperature where the two balance.\n",{"path":17051,"title":17052,"module":17048,"summary":17053},"\u002Fstatistical-mechanics\u002Finteractions\u002Fthe-van-der-waals-gas-and-liquid-gas-coexistence","The van der Waals Gas and Liquid-Gas Coexistence","Resumming the second virial coefficient $B_2=b-a\u002Fk_BT$ into an equation of state gives the van der Waals model $(P+a\u002Fv^2)(v-b)=k_BT$, the simplest theory of a fluid that condenses. Below the critical temperature its isotherms develop a mechanically unstable loop; the Maxwell equal-area construction replaces the loop with a coexistence tie line. The critical point sits at $v_c=3b$, $k_BT_c=8a\u002F27b$, $P_c=a\u002F27b^2$, and the model predicts universal but incorrect critical exponents because it ignores fluctuations.\n",{"path":17055,"title":17056,"module":17048,"summary":17057},"\u002Fstatistical-mechanics\u002Finteractions\u002Fquantum-gases-with-interactions-and-exchange","Quantum Gases with Interactions and Statistical Exchange","A quantum gas has a nonzero second virial coefficient even with no forces between the particles: symmetrization alone produces an effective statistical interaction, attractive for bosons and repulsive for fermions, with range the thermal wavelength $\\lambda$. This lesson derives that exchange contribution $B_2=\\mp\\lambda^3\u002F2^{5\u002F2}g$, writes it as a statistical potential $v_s(r)=-k_BT\\ln(1\\pm e^{-2\\pi r^2\u002F\\lambda^2})$, and shows how real interactions add on top through the Beth-Uhlenbeck phase-shift formula, reducing at low temperature to a single scattering length.\n",{"path":17059,"title":17060,"module":17061,"summary":17062},"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fphases-coexistence-and-classification","Phases, Coexistence, and the Classification of Transitions","Phase Transitions","A phase transition is a point where the free energy of a substance loses analyticity, so a small change in temperature or pressure produces a qualitative change of state. This lesson maps the coexistence curves of a pure substance, derives the Clausius-Clapeyron relation between the slope of a coexistence line and its latent heat, and separates first-order transitions (discontinuous entropy and density) from continuous ones (a vanishing order parameter and divergent response). The Ehrenfest scheme, the order parameter, and the triple and critical points fix the vocabulary the rest of the module builds on.\n",{"path":17064,"title":17065,"module":17061,"summary":17066},"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fthe-ising-model-and-exact-solutions","The Ising Model and Exact Results","The Ising model reduces cooperative ordering to spins on a lattice coupled to their neighbors, and the same Hamiltonian describes uniaxial magnets, the liquid-gas critical point through the lattice gas, and binary alloys. This lesson solves the one-dimensional chain exactly with the transfer matrix, shows by a domain-wall argument why one dimension has no ordered phase at any positive temperature, contrasts the survival of order in two dimensions, and quotes Onsager's exact two-dimensional results: the critical temperature, the logarithmically divergent heat capacity, and the magnetization exponent one eighth.\n",{"path":17068,"title":17069,"module":17061,"summary":17070},"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fmean-field-theory-and-the-weiss-model","Mean-Field Theory and Spontaneous Symmetry Breaking","Mean-field theory replaces the neighbors of each spin by their average, turning the interacting Ising model into a single spin in a self-consistent field. The resulting equation m = tanh(beta J z m + beta h) has only the zero solution above a critical temperature and gains a nonzero root below it, giving spontaneous magnetization and a mean-field critical temperature k T_c = J z. The Bragg-Williams free energy turns single-welled above T_c and double-welled below, the picture of spontaneous symmetry breaking. The approximation is exact in high dimension and fails below the upper critical dimension four, quantified by the Ginzburg criterion.\n",{"path":17072,"title":17073,"module":17061,"summary":17074},"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fcritical-exponents-and-landau-theory","Critical Exponents, Scaling, and Landau Theory","Near a continuous transition every singular quantity follows a power law in the reduced temperature, and the exponents alpha, beta, gamma, delta, nu, and eta encode the transition more sharply than T_c itself. Landau theory expands the free energy in the order parameter and delivers the mean-field exponents in a few lines. They disagree with experiment and with the exact two-dimensional Ising values, but the exponents are not independent: the scaling relations of Rushbrooke, Widom, Fisher, and Josephson tie them together, and the correlation length sets the length scale that organizes universality classes.\n",{"path":17076,"title":17077,"module":17061,"summary":17078},"\u002Fstatistical-mechanics\u002Fphase-transitions\u002Fthe-renormalization-group-idea","Scaling and the Renormalization-Group Idea","At a critical point fluctuations exist on every length scale, so the system looks the same after coarse-graining. The renormalization group makes this self-similarity a computation: group spins into blocks, integrate out the short scales, and track how the couplings change. The transformation has fixed points, and the flow near a critical fixed point separates relevant couplings that grow from irrelevant ones that shrink, which is why only dimension and symmetry survive to set the exponents. The one-dimensional Ising decimation carries the whole scheme through in closed form and reproduces the absence of a finite-temperature transition.\n",{"path":17080,"title":17081,"module":17082,"summary":17083},"\u002Fstatistical-mechanics\u002Ffluctuations\u002Fthermodynamic-fluctuations-and-response","Thermodynamic Fluctuations and Response Functions","Fluctuations and Response","Thermodynamic variables are sharp only on average; a macroscopic system in equilibrium fluctuates about its mean values. Einstein inverted Boltzmann's $S=k_B\\ln\\Omega$ into a Gaussian probability for a fluctuation, $w\\propto e^{\\Delta S\u002Fk_B}$, and the second moments it predicts reproduce the response functions: $\\langle\\Delta E^2\\rangle=k_BT^2C_V$, $\\langle\\Delta V^2\\rangle=k_BTV\\kappa_T$, $\\langle\\Delta M^2\\rangle=k_BT\\chi_T$. The variances diverge where the responses diverge, at a critical point, producing critical opalescence and the breakdown of the thermodynamic description.\n",{"path":17085,"title":17086,"module":17082,"summary":17087},"\u002Fstatistical-mechanics\u002Ffluctuations\u002Fbrownian-motion-and-the-langevin-equation","Brownian Motion and the Langevin Equation","A pollen grain in water executes a random walk driven by molecular collisions. Einstein tied its diffusion constant to its mobility, $D=\\mu_{\\mathrm{mob}}k_BT$, turning a visible motion into a measurement of Avogadro's number. The Langevin equation splits the collisions into a systematic drag and a random force whose strength is fixed by the drag through $\\langle\\xi(t)\\xi(t')\\rangle=2\\gamma k_BT\\,\\delta(t-t')$ — the first fluctuation–dissipation relation. The mean-square displacement grows ballistically at short times and linearly, $\\langle r^2\\rangle=2dDt$, at long times, and the Stokes–Einstein relation $D=k_BT\u002F6\\pi\\eta a$ closes the loop to Perrin's experiments.\n",{"path":17089,"title":17090,"module":17082,"summary":17091},"\u002Fstatistical-mechanics\u002Ffluctuations\u002Flinear-response-and-the-fluctuation-dissipation-theorem","Linear Response and the Fluctuation-Dissipation Theorem","A system driven by a weak external field responds through a generalized susceptibility $\\chi(\\omega)$ whose imaginary part measures dissipation. The Wiener–Khinchin theorem makes the power spectrum of equilibrium fluctuations the Fourier transform of their correlation function, and the fluctuation–dissipation theorem ties the two together: $S_x(\\omega)=(2k_BT\u002F\\omega)\\,\\chi''(\\omega)$, so the spectrum of spontaneous fluctuations is fixed by the dissipative response. The Johnson–Nyquist noise of a resistor, $\\langle V^2\\rangle=4k_BTR\\,\\Delta f$, is the canonical example, and Onsager reciprocity closes the subject.\n",{"path":17093,"title":17094,"module":6,"summary":6},"\u002Fstatistical-mechanics","Statistical Mechanics",{"path":17096,"title":17097,"module":17098,"summary":17099},"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fbonding-mechanisms","Bonding Mechanisms","Molecules and Chemical Bonding","A molecule forms when the total energy of two atoms drops below the energy of the separated pair. This lesson works through the four mechanisms that produce that minimum: the ionic bond from charge transfer, the covalent bond from shared electron wave functions, the metallic bond, and the weak dipole-dipole and hydrogen bonds, computing bond lengths and dissociation energies for NaCl, H₂, and H₂⁺.\n",{"path":17101,"title":17102,"module":17098,"summary":17103},"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fmolecular-orbitals-and-h2-plus","The Molecular-Orbital Method and H₂⁺","The hydrogen molecule ion is the two-center problem that fixes the language of chemical bonding. This lesson builds the molecular orbital as a linear combination of atomic orbitals, minimizes the energy through the variational secular equation, and reduces the result to three two-center integrals: the overlap, the Coulomb term, and the exchange (resonance) integral. The bonding and antibonding levels, their potential-energy curves, and the charge piled between the nuclei follow from those integrals.\n",{"path":17105,"title":17106,"module":17098,"summary":17107},"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fhydrogen-molecule-and-exchange","The Hydrogen Molecule, Exchange, and Hybridization","Adding the second electron turns the one-electron ion into the two-electron hydrogen molecule, where electron-electron repulsion and the Pauli principle govern the bond. This lesson contrasts the Heitler-London valence-bond and molecular-orbital wave functions, derives the singlet-triplet splitting as an exchange energy, shows why naive molecular orbitals fail at dissociation, and builds the sp, sp², and sp³ hybrids that fix the directed geometry of covalent bonds.\n",{"path":17109,"title":17110,"module":17098,"summary":17111},"\u002Fcondensed-matter\u002Fmolecules-and-bonding\u002Fvan-der-waals-forces","Van der Waals Forces","The bond of last resort acts between all atoms, even closed-shell noble gases with no permanent moment. This lesson separates the three van der Waals contributions — Keesom orientation, Debye induction, and London dispersion — derives the London 1\u002Fr⁶ attraction from the coupled-oscillator and second-order perturbation pictures, and assembles the Lennard-Jones potential to compute the equilibrium spacing and cohesive energy of the noble-gas crystals, argon in particular.\n",{"path":17113,"title":17114,"module":17115,"summary":17116},"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Frotational-vibrational-spectra","Rotational and Vibrational Spectra of Molecules","Molecular Spectra","A diatomic molecule stores energy in three well-separated ledgers: electronic, vibrational, and rotational. Quantizing the rigid rotor gives levels spaced as ℓ(ℓ+1); quantizing the bond as a harmonic oscillator gives equally spaced vibrational levels. Their combination produces the P and R branches of an infrared absorption band, from which the bond length and force constant are read directly.\n",{"path":17118,"title":17119,"module":17115,"summary":17120},"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Fanharmonicity-and-rovibrational-structure","Anharmonicity and Rovibrational Structure","The rigid rotor and harmonic oscillator are first approximations. A real bond follows the Morse potential, whose levels converge toward dissociation; a real rotor stretches centrifugally; and vibration couples to rotation, so the rotational constant depends on the vibrational level. This lesson works out the anharmonic and centrifugal corrections, the Birge-Sponer route to the dissociation energy, the isotope shift, and the thermal band envelope.\n",{"path":17122,"title":17123,"module":17115,"summary":17124},"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Framan-and-electronic-bands","Raman Scattering and Electronic Bands","Not every vibration absorbs in the infrared. Raman scattering reaches modes that modulate the polarizability, giving Stokes and anti-Stokes lines whose intensity ratio measures temperature, and the mutual-exclusion rule pairs it with infrared absorption. Electronic transitions add the vibronic structure of band spectra, governed by the Franck-Condon principle, and the radiative fates of an excited state are sorted by the Jablonski diagram into fluorescence and phosphorescence.\n",{"path":17126,"title":17127,"module":17115,"summary":17128},"\u002Fcondensed-matter\u002Fmolecular-spectra\u002Flasers-and-masers","Lasers, Masers, and Stimulated Emission","Einstein's three radiative processes — absorption, spontaneous emission, and stimulated emission — and the coefficients that relate them. Stimulated emission produces coherent photons, and inverting the level populations turns it into net amplification. We build the ruby three-level laser and the helium-neon four-level laser, and show why the fourth level makes inversion easy.\n",{"path":17130,"title":17131,"module":17132,"summary":17133},"\u002Fcondensed-matter\u002Fcrystal-structure\u002Fstructure-of-solids","The Structure of Solids","Crystal Structure","A crystal is a unit cell repeated in three dimensions. We classify the common cubic lattices, compute the Coulomb energy of an ionic crystal through the Madelung constant, and show how the divergent naive lattice sum is tamed by cubic shells. The cohesive energy that results predicts melting points and connects the diatomic bond of an earlier lesson to the bulk solid.\n",{"path":17135,"title":17136,"module":17132,"summary":17137},"\u002Fcondensed-matter\u002Fcrystal-structure\u002Fbravais-lattices-and-crystal-systems","Bravais Lattices, Bases, and Crystal Structures","A crystal is a Bravais lattice decorated by a basis. This lesson separates the two, builds primitive and Wigner-Seitz cells, enumerates the seven crystal systems and fourteen Bravais lattices, and fixes the language of point and space groups. Miller indices label planes and directions, and the packing fractions of the close-packed, cubic, and diamond structures follow from the geometry.\n",{"path":17139,"title":17140,"module":17132,"summary":17141},"\u002Fcondensed-matter\u002Fcrystal-structure\u002Freciprocal-lattice-and-brillouin-zones","The Reciprocal Lattice and Brillouin Zones","Every periodic crystal has a dual lattice in wavevector space. This lesson defines the reciprocal lattice through the condition b_i dot a_j equals two pi delta, derives its properties, shows the reciprocal of fcc is bcc, links reciprocal vectors to families of lattice planes, and builds the first Brillouin zone as the Wigner-Seitz cell of the reciprocal lattice, including the higher zones.\n",{"path":17143,"title":17144,"module":17132,"summary":17145},"\u002Fcondensed-matter\u002Fcrystal-structure\u002Fdiffraction-and-structure-factors","X-ray and Neutron Diffraction","A crystal diffracts radiation whose wavelength matches its atomic spacing. This lesson derives the Bragg condition, the equivalent Laue condition 2k dot G equals G squared, and the Ewald-sphere construction, then computes the geometric structure factor that produces systematic absences for bcc and fcc, the atomic form factor, and the powder method. It closes on why neutrons and electrons complement X-rays.\n",{"path":17147,"title":17148,"module":17149,"summary":17150},"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fphonon-dispersion","The Harmonic Crystal and Phonon Dispersion","Lattice Dynamics","Atoms in a crystal vibrate about their equilibrium sites, and expanding the potential to second order turns the whole lattice into a set of coupled harmonic oscillators. This lesson sets up the harmonic approximation and the dynamical matrix, solves the monatomic linear chain for its dispersion omega(k) = 2 sqrt(K\u002FM) times the absolute sine of ka over two, explains why wavevectors outside the first Brillouin zone are redundant, and extends the chain to two atoms per cell to produce acoustic and optical branches with a frequency gap.\n",{"path":17152,"title":17153,"module":17149,"summary":17154},"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fphonons-quantization-and-dos","Phonons, Density of States, and Crystal Momentum","Quantizing the normal modes of a harmonic crystal turns each vibrational mode into a quantum oscillator whose excitations are phonons. This lesson counts phonons with Bose-Einstein statistics, defines crystal momentum and the normal versus Umklapp distinction in momentum conservation, builds the density of states with its van Hove singularities, and shows how inelastic neutron scattering measures a dispersion curve point by point.\n",{"path":17156,"title":17157,"module":17149,"summary":17158},"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fdebye-einstein-heat-capacity","Thermal Properties — Einstein and Debye Models","The lattice heat capacity follows the classical Dulong-Petit value at high temperature but collapses toward zero as T approaches zero, a purely quantum effect. This lesson derives that behavior from the Einstein model of a single frequency, then the Debye model of a linear phonon spectrum with a cutoff, obtaining the Debye T-cubed law at low temperature and the Debye interpolation across all temperatures, and closes with thermal expansion and the Gruneisen parameter.\n",{"path":17160,"title":17161,"module":17149,"summary":17162},"\u002Fcondensed-matter\u002Flattice-dynamics\u002Fanharmonicity-and-thermal-transport","Anharmonicity, Thermal Expansion, and Heat Conduction","A perfectly harmonic crystal neither expands when heated nor resists heat flow. Both effects come from the cubic and higher terms the harmonic approximation discards. This lesson derives thermal expansion from an asymmetric interatomic potential, treats phonon-phonon scattering as the decay channel these terms open, shows why Umklapp processes are what make lattice thermal conductivity finite, and traces the temperature dependence of the conductivity and the phonon mean free path.\n",{"path":17164,"title":17165,"module":17166,"summary":17167},"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Ffree-electron-gas-and-conduction","Conduction and the Free-Electron Gas","Free-Electron Fermi Gas","Drude's classical free-electron model gets Ohm's law right but the resistivity, its temperature dependence, and the heat capacity wrong. Replacing the Maxwell-Boltzmann distribution with the Fermi-Dirac distribution and treating electron-lattice collisions as wave scattering repairs all three: the Fermi energy, Fermi speed, and a mean free path set by thermal lattice vibrations.\n",{"path":17169,"title":17170,"module":17166,"summary":17171},"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Fsommerfeld-model-and-heat-capacity","The Sommerfeld Model: Ground State and Heat Capacity","Quantizing the free-electron gas in a box fills a Fermi sphere in k-space. The density of states grows as the square root of energy in three dimensions, and the Fermi energy, temperature, and wavevector follow for real metals. The Sommerfeld expansion shows only a thermal shell of width k_BT near E_F is excited, giving an electronic heat capacity linear in T that sits beneath the phonon T-cubed term.\n",{"path":17173,"title":17174,"module":17166,"summary":17175},"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Ftransport-and-the-hall-effect","Transport, Wiedemann–Franz, and the Hall Effect","The relaxation-time picture displaces the Fermi sphere under an applied field and gives the electrical conductivity ne-squared-tau over m. The same electrons carry heat, and their ratio yields the Wiedemann–Franz law with the universal Lorenz number. A magnetic field bends the carriers into cyclotron orbits and produces the Hall voltage, whose sign reveals the charge of the carriers.\n",{"path":17177,"title":17178,"module":17166,"summary":17179},"\u002Fcondensed-matter\u002Ffree-electron-fermi-gas\u002Fscreening-and-plasmons","Screening, Plasmons, and the Limits of Free Electrons","A mobile electron gas rearranges to screen any foreign charge, turning the bare Coulomb potential into a short-ranged Yukawa form over the Thomas–Fermi length. Displaced collectively, the gas rings at the plasma frequency, whose quantum is the plasmon and whose value sets the reflectivity edge of metals. A ledger of free-electron successes and failures then motivates band theory.\n",{"path":17181,"title":17182,"module":17183,"summary":17184},"\u002Fcondensed-matter\u002Fband-theory\u002Fblochs-theorem-and-energy-bands","Bloch's Theorem and Energy Bands","Band Theory","An electron in a periodic potential has stationary states that are plane waves modulated by a lattice-periodic envelope. This lesson proves Bloch's theorem two ways, defines crystal momentum and the band index, counts the allowed wavevectors from Born–von Kármán boundary conditions, and sets up the extended, reduced, and repeated-zone descriptions of a band.\n",{"path":17186,"title":17187,"module":17183,"summary":17188},"\u002Fcondensed-matter\u002Fband-theory\u002Fnearly-free-electron-model","The Nearly-Free-Electron Model","A weak periodic potential leaves the free-electron parabola almost intact except near Brillouin-zone boundaries, where two nearly degenerate plane waves mix. This lesson solves the resulting two-by-two secular problem, shows the gap of size twice the potential component opening at each boundary, identifies the two standing waves that pile charge on and between the ions, and works the exactly solvable Kronig–Penney model.\n",{"path":17190,"title":17191,"module":17183,"summary":17192},"\u002Fcondensed-matter\u002Fband-theory\u002Ftight-binding-method","The Tight-Binding Method","The opposite limit to nearly-free electrons builds bands from atomic orbitals. A Bloch sum of one orbital per site gives a dispersion set by the hopping integral between neighbours; the band widens from a sharp atomic level as the atoms approach. This lesson derives the s-band cosine dispersion, extends it to p-bands, and introduces Wannier functions as the localized dual of Bloch states.\n",{"path":17194,"title":17195,"module":17183,"summary":17196},"\u002Fcondensed-matter\u002Fband-theory\u002Ffermi-surfaces-and-semiclassical-dynamics","Fermi Surfaces, Effective Mass, and Metals vs Insulators","Filling the bands settles which crystals conduct. A filled band carries no current, so a crystal with filled bands and a gap is an insulator, while a partly filled band makes a metal. This lesson derives the no-current theorem for a filled band, defines the Fermi surface and Harrison's construction, introduces holes and the effective mass from band curvature, and states the semiclassical equations of motion that lead to Bloch oscillations.\n",{"path":17198,"title":17199,"module":17200,"summary":17201},"\u002Fcondensed-matter\u002Fsemiconductors\u002Fsemiconductor-bands-and-junctions","Band Theory and Semiconductors","Semiconductors","The periodic lattice splits atomic levels into allowed energy bands separated by forbidden gaps. Whether the highest occupied band is full or partly full, and how wide the gap above it is, sorts every solid into conductor, insulator, or semiconductor. Doping adds donor or acceptor levels inside the gap, and a p-n junction built from doped regions gives the diode, the solar cell, the LED, and the transistor.\n",{"path":17203,"title":17204,"module":17200,"summary":17205},"\u002Fcondensed-matter\u002Fsemiconductors\u002Fintrinsic-and-extrinsic-semiconductors","Carrier Statistics: Intrinsic and Extrinsic Semiconductors","The number of mobile electrons and holes in a semiconductor follows from the density of states near each band edge and the Fermi-Dirac tail that reaches into it. This lesson derives the effective densities of states, the intrinsic concentration and its exponential gap dependence, the law of mass action, the temperature march of the Fermi level, and the freeze-out, saturation, and intrinsic regimes of a doped crystal.\n",{"path":17207,"title":17208,"module":17200,"summary":17209},"\u002Fcondensed-matter\u002Fsemiconductors\u002Fcarrier-transport-and-recombination","Carrier Transport and Recombination","Carriers move by drift in a field and by diffusion down a concentration gradient, the two tied together by the Einstein relation. This lesson derives mobility and its scattering-limited temperature dependence, the drift and diffusion currents, the continuity equations, band-to-band and trap-assisted recombination, and the minority-carrier lifetime and diffusion length that set the length scale of every junction device.\n",{"path":17211,"title":17212,"module":17200,"summary":17213},"\u002Fcondensed-matter\u002Fsemiconductors\u002Fthe-pn-junction","The p-n Junction in Depth","Joining p-type and n-type silicon aligns their Fermi levels and leaves a depletion region of fixed charge with a built-in potential. This lesson derives the space-charge field and potential from Poisson's equation in the depletion approximation, the built-in voltage from Fermi-level alignment, the Shockley diode equation from minority-carrier diffusion, junction and diffusion capacitance, and the avalanche and Zener breakdown mechanisms.\n",{"path":17215,"title":17216,"module":17200,"summary":17217},"\u002Fcondensed-matter\u002Fsemiconductors\u002Ftransistors-and-optoelectronics","Transistors and Optoelectronic Devices","Two junctions in series make a bipolar transistor whose thin base gives current gain; a gate over an oxide makes a MOSFET whose inversion channel switches digital logic. Run in reverse, a junction converts photons to current. This lesson derives the transistor current gain and the MOSFET channel current, then treats the LED, the diode laser, and the illuminated solar-cell characteristic.\n",{"path":17219,"title":17220,"module":17221,"summary":17222},"\u002Fcondensed-matter\u002Fdielectrics-and-ferroelectrics\u002Fdielectrics-and-polarization","Dielectrics, Polarization, and the Local Field","Dielectrics and Ferroelectrics","An insulator responds to an electric field by polarizing. This lesson builds the macroscopic polarization and the dielectric constant, sorts the microscopic response into electronic, ionic, and orientational polarizability, and corrects the field an atom actually feels to the Lorentz local field E + P\u002F3 epsilon-0. The Clausius-Mossotti relation links the measured permittivity to the atomic polarizability, and the frequency dependence of each mechanism explains why the static and optical dielectric constants differ.\n",{"path":17224,"title":17225,"module":17221,"summary":17226},"\u002Fcondensed-matter\u002Fdielectrics-and-ferroelectrics\u002Fferroelectrics-and-piezoelectrics","Ferroelectrics, Piezoelectrics, and Structural Transitions","Some crystals carry a polarization with no applied field and switch it under a reversing field, tracing a hysteresis loop. This lesson develops the ferroelectric transition through the perovskite BaTiO3 displacive instability and its soft transverse-optical mode, builds the Landau free-energy theory of first- and second-order polar transitions, derives the Curie-Weiss divergence of the dielectric constant, and closes with piezoelectricity and pyroelectricity and their devices.\n",{"path":17228,"title":17229,"module":17230,"summary":17231},"\u002Fcondensed-matter\u002Fmagnetism\u002Fdiamagnetism-and-paramagnetism","Diamagnetism and Paramagnetism","Magnetism in Solids","Every solid responds to a magnetic field. Filled shells give a small negative diamagnetic susceptibility from induced Larmor currents; localized moments give a positive Curie paramagnetism described by the Brillouin function, with the ground-state moment fixed by Hund's rules. The conduction electrons add a temperature-independent Pauli paramagnetism from the thermal shell near the Fermi surface, partly cancelled by Landau diamagnetism of their orbital motion.\n",{"path":17233,"title":17234,"module":17230,"summary":17235},"\u002Fcondensed-matter\u002Fmagnetism\u002Fexchange-and-ferromagnetism","Exchange and Ferromagnetism","Magnetic ordering at hundreds of kelvin cannot be dipolar; it is an exchange effect, the Coulomb repulsion sorted by the Pauli principle into a spin-dependent energy captured by the Heisenberg Hamiltonian. Weiss molecular-field theory replaces the exchange field by an average proportional to the magnetization, giving a self-consistent equation whose solution is spontaneous magnetization below a Curie temperature and a Curie–Weiss susceptibility above it. Itinerant ferromagnetism follows from the Stoner criterion on the band density of states.\n",{"path":17237,"title":17238,"module":17230,"summary":17239},"\u002Fcondensed-matter\u002Fmagnetism\u002Fantiferromagnetism-and-domains","Antiferromagnetism, Ferrimagnetism, and Domains","A negative exchange coupling orders neighboring spins antiparallel. Two-sublattice molecular-field theory gives a Néel temperature marked by a cusp in the susceptibility, and unequal sublattices leave a net moment — ferrimagnetism, the magnetism of the ferrites. A ferromagnet breaks into domains to reduce its magnetostatic energy, separated by Bloch walls whose width is set by the competition between exchange and magnetocrystalline anisotropy, and the irreversible motion of those walls produces the hysteresis loop.\n",{"path":17241,"title":17242,"module":17230,"summary":17243},"\u002Fcondensed-matter\u002Fmagnetism\u002Fspin-waves-and-magnons","Spin Waves and Magnons","The lowest excitations of a ferromagnet are not single flipped spins but collective precessions in which every moment tips slightly and its phase advances along the crystal. These spin waves have a quadratic dispersion at long wavelength, quantize into magnons obeying Bose statistics, and their thermal population removes magnetization as the Bloch T-to-the-three-halves law. Antiferromagnetic magnons disperse linearly, and inelastic neutron scattering measures both.\n",{"path":17245,"title":17246,"module":17247,"summary":17248},"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fsuperconductivity-phenomenology","Superconductivity: Phenomenology and BCS","Superconductivity","Below a critical temperature some materials lose all resistance and expel magnetic flux — the Meissner effect that defines the state. The isotope effect points to lattice vibrations, and BCS theory binds electrons into Cooper pairs through phonon exchange. The paired condensate opens an energy gap, quantizes magnetic flux, and drives the Josephson effects.\n",{"path":17250,"title":17251,"module":17247,"summary":17252},"\u002Fcondensed-matter\u002Fsuperconductivity\u002Flondon-theory-and-the-meissner-effect","London Theory and the Meissner Effect","A perfect conductor freezes the field it was cooled in; a superconductor expels it. The distinction needs a constitutive law beyond zero resistance — the two London equations — whose solution is exponential flux decay over the penetration depth. The same rigidity follows from a macroscopic condensate wave function, and the thermodynamics of the critical field fixes the condensation energy, the latent heat, and the specific-heat jump.\n",{"path":17254,"title":17255,"module":17247,"summary":17256},"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fginzburg-landau-theory","Ginzburg–Landau Theory, Vortices, and Type-II","A complex order parameter and a free-energy expansion turn the superconducting transition into a Landau theory. Two lengths emerge — the coherence length and the penetration depth — whose ratio kappa sorts superconductors into type I and type II. Type-II materials admit flux as an Abrikosov lattice of vortices, each threading exactly one quantum h\u002F2e, between a lower and an upper critical field.\n",{"path":17258,"title":17259,"module":17247,"summary":17260},"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fbcs-theory","Microscopic BCS Theory","A phonon-mediated attraction, however weak, binds two electrons above the Fermi sea — the Cooper problem shows the sea is unstable. The BCS variational ground state pairs all electrons near the Fermi surface and, through a self-consistent gap equation, opens an energy gap. Weak-coupling solution gives the exponential T_c and the universal ratios 2 Delta(0) = 3.53 k_B T_c and Delta C \u002F C_n = 1.43.\n",{"path":17262,"title":17263,"module":17247,"summary":17264},"\u002Fcondensed-matter\u002Fsuperconductivity\u002Fjosephson-and-high-tc","Josephson Effects and Unconventional Superconductors","Two superconductors joined by a thin barrier carry a supercurrent set by their phase difference — the dc Josephson effect — and oscillate at 2eV\u002Fh under a voltage. A two-junction loop turns flux quantization into a magnetometer of single-quantum sensitivity. The cuprates superconduct in CuO2 planes with a doping-dependent dome, d-wave pairing, and a pseudogap that lie outside the phonon picture.\n",{"path":17266,"title":17267,"module":17268,"summary":17269},"\u002Fcondensed-matter\u002Fnanostructures\u002Fquantum-wells-wires-and-dots","Quantum Wells, Wires, and Dots","Nanostructures","When a crystal is shrunk until one or more of its dimensions approaches the electron wavelength, the continuous bands of the bulk break into discrete subbands. Confining in one direction gives a quantum well with a step-like density of states, in two directions a quantum wire with inverse-square-root singularities, and in all three a quantum dot whose levels are sharp like an atom's. This lesson derives the density of states in each case and applies it to size-tunable dot emission and the Coulomb blockade of a single-electron transistor.\n",{"path":17271,"title":17272,"module":17268,"summary":17273},"\u002Fcondensed-matter\u002Fnanostructures\u002Finteger-quantum-hall-effect","The 2D Electron Gas and the Integer Quantum Hall Effect","A two-dimensional electron gas in a strong perpendicular magnetic field has its continuous density of states collapse into macroscopically degenerate Landau levels. As the field is swept, the Hall resistance locks onto exact plateaus at h over an integer times e squared, while the longitudinal resistance drops to zero. This lesson derives the Landau levels and their degeneracy, explains the plateaus through disorder-localized states and current-carrying edge channels, and states why the von Klitzing constant is now a resistance standard.\n",{"path":17275,"title":17276,"module":17268,"summary":17277},"\u002Fcondensed-matter\u002Fnanostructures\u002Ffractional-quantum-hall-and-topology","The Fractional Quantum Hall Effect and Topological Order","When the lowest Landau level is only partly filled, the non-interacting theory predicts no gap, yet a plateau appears at filling one-third. It is a many-body effect: Coulomb repulsion selects a correlated ground state, the Laughlin wavefunction, whose excitations carry a fraction of the electron charge. This lesson builds the Laughlin state, introduces composite fermions that map the fractional effect onto an integer one, and explains how the quantum Hall effect brought the Chern number and topology into condensed-matter physics.\n",{"path":17279,"title":17280,"module":17268,"summary":17281},"\u002Fcondensed-matter\u002Fnanostructures\u002Fgraphene-and-dirac-materials","Graphene and Dirac Materials","Graphene is one atomic layer of carbon on a honeycomb lattice. A tight-binding calculation on its two-atom basis gives valence and conduction bands that touch at the corners of the Brillouin zone, where the dispersion is linear and the electrons behave as massless two-dimensional Dirac particles with a fixed speed. This lesson derives the Dirac cones, the Berry phase of pi and the sublattice chirality, the anomalous half-integer quantum Hall effect that follows, and how opening a gap in a Dirac cone points toward topological insulators.\n",{"path":17283,"title":17284,"module":6,"summary":6},"\u002Fcondensed-matter","Condensed Matter Physics",{"path":17286,"title":17287,"module":14054,"summary":17288},"\u002Flogic\u002Ffoundations\u002Flogic-as-a-mathematical-model","Logic as a Mathematical Model of Deduction","Symbolic logic models deductive reasoning the way probability theory models chance: it keeps the form of a correct deduction and discards its content. A deduction is valid when its conclusion follows from the form of the premises alone, independent of what the non-logical words mean. Two models carry the subject — coarse sentential logic and fine first-order logic — and four questions organize it: logical consequence, methods of proof, the gap between provable and true, and the link between logic and computability. Tuples, relations, functions, equivalence classes, and cardinality supply the set-theoretic vocabulary every later chapter uses.\n",{"path":17290,"title":17291,"module":17292,"summary":17293},"\u002Flogic\u002Fsentential-logic\u002Fformal-languages-and-well-formed-formulas","Formal Languages and Well-Formed Formulas","Sentential Logic","The language of sentential logic has an alphabet of sentence symbols, five connectives, and two parentheses, with formation rules that pick out the well-formed formulas. The wffs are the least set of expressions closed under the five formula-building operations, and every such generated set carries an induction principle.\n",{"path":17295,"title":17296,"module":17292,"summary":17297},"\u002Flogic\u002Fsentential-logic\u002Ftruth-assignments-and-tautologies","Truth Assignments, Tautologies, and Consequence","A truth assignment fixes the sentence symbols true or false, and a recursion extends it uniquely to every formula. Satisfaction, tautologies, and tautological implication — one formula following semantically from others — rest on that extension, and the truth-table procedure decides implication for finite premise sets.\n",{"path":17299,"title":17300,"module":17292,"summary":17301},"\u002Flogic\u002Fsentential-logic\u002Funique-readability-and-parsing","Unique Readability and a Parsing Algorithm","Parentheses keep a formula from being read two ways. The parenthesis lemmas and a top-down parsing algorithm recover a formula's structure and yield unique readability: every wff has exactly one formation tree, which is what makes the truth recursion well defined.\n",{"path":17303,"title":17304,"module":17292,"summary":17305},"\u002Flogic\u002Fsentential-logic\u002Finduction-and-recursion","Induction and Recursion on Formulas","Two principles govern any set generated from initial elements by operations: prove a property of all its members by covering the initial elements and the closure steps, and define a function on it by recursion on structure. The recursion theorem needs the set to be freely generated, and unique readability supplies that condition for the well-formed formulas.\n",{"path":17307,"title":17308,"module":17292,"summary":17309},"\u002Flogic\u002Fsentential-logic\u002Fexpressive-completeness-and-normal-forms","Sentential Connectives and Normal Forms","Every formula computes a Boolean function of its atoms, and Post's theorem gives the converse: every Boolean function is realized by a wff in disjunctive normal form, so the five connectives are more than enough. Minimal complete sets follow, down to the single connectives NAND and NOR, together with a method for proving a set of connectives incomplete.\n",{"path":17311,"title":17312,"module":17292,"summary":17313},"\u002Flogic\u002Fsentential-logic\u002Fboolean-circuits","Switching Circuits","A memoryless two-valued circuit computes a Boolean function, so every formula names a gate network and every network a formula. Cost and delay are read off the formula by recursion, and tautological equivalence and normal forms design and simplify circuits realizing a given specification.\n",{"path":17315,"title":17316,"module":17292,"summary":17317},"\u002Flogic\u002Fsentential-logic\u002Fcompactness-and-effectiveness","Compactness and Effectiveness","The compactness theorem reduces satisfiability of an infinite set of formulas to its finite subsets, proved by extension to a maximal finitely satisfiable set and applied to color infinite graphs. Effectiveness fixes what \"decidable\" and \"effectively enumerable\" mean and settles the decidability of tautologyhood.\n",{"path":17319,"title":17320,"module":17321,"summary":17322},"\u002Flogic\u002Ffirst-order-languages\u002Ffirst-order-languages","First-Order Languages","First-Order Languages and Structures","Sentential logic cannot see inside a simple statement, so it misses valid arguments that turn on quantifiers and predicates. A first-order language adds a quantifier, variables, and a chosen vocabulary of predicate, function, and constant symbols. Terms and well-formed formulas are built by recursion over this alphabet, and a variable occurs free or bound according to the quantifiers that reach it.\n",{"path":17324,"title":17325,"module":17321,"summary":17326},"\u002Flogic\u002Ffirst-order-languages\u002Fstructures-truth-and-satisfaction","Structures, Truth, and Satisfaction","A structure interprets a language: a nonempty universe plus a meaning for every predicate, function, and constant symbol. Tarski's recursion defines when a structure satisfies a formula under a variable assignment, and hence when a sentence is true. From satisfaction we recover logical implication, validity, and logical equivalence for first-order logic.\n",{"path":17328,"title":17329,"module":17321,"summary":17330},"\u002Flogic\u002Ffirst-order-languages\u002Fdefinability-and-elementary-equivalence","Definability and Elementary Equivalence","Fix a structure and ask which relations a formula can pick out: the definable ones. A set of sentences picks out a class of structures, the elementary classes. Homomorphisms and isomorphisms compare structures, and the homomorphism theorem shows isomorphic structures satisfy the same sentences. Automorphisms bound what first-order logic can distinguish, giving a tool for proving relations undefinable.\n",{"path":17332,"title":17333,"module":17321,"summary":17334},"\u002Flogic\u002Ffirst-order-languages\u002Fterms-substitution-and-parsing","Parsing, Substitution, and Substitutability","Every recursion on first-order syntax rests on unique readability. A parenthesis-counting function proves that terms and formulas decompose in exactly one way, and a parsing algorithm recovers the decomposition. Substituting a term for a free variable can capture it under a quantifier; the substitutability condition rules that out, and the substitution lemma trades syntactic substitution for a change of assignment.\n",{"path":17336,"title":17337,"module":17338,"summary":17339},"\u002Flogic\u002Fdeductive-calculus\u002Fa-deductive-calculus","A Deductive Calculus for First-Order Logic","The Deductive Calculus and Its Metatheorems","A proof must be finite and mechanically checkable. A Hilbert-style calculus meets both demands: six schemas of logical axioms, a single rule of inference (modus ponens), and the syntactic consequence relation they generate. Substitution and substitutability are defined by recursion, and the bridge theorem reduces deducibility to tautological implication from the axioms.\n",{"path":17341,"title":17342,"module":17338,"summary":17343},"\u002Flogic\u002Fdeductive-calculus\u002Fdeduction-theorem-and-derived-rules","The Deduction Theorem and Derived Rules","Raw deductions from axioms are unusable by hand. The generalization theorem, the deduction theorem, contraposition, reductio ad absurdum, and rule T reduce the calculus to the moves of ordinary mathematics, each proved once to license a block of axiom-level steps. Generalization on constants and alphabetic variants handle the quantifier and substitution bookkeeping.\n",{"path":17345,"title":17346,"module":17338,"summary":17347},"\u002Flogic\u002Fdeductive-calculus\u002Fsoundness","The Soundness Theorem","Soundness is the easy half of the match between proof and truth. Whatever the calculus deduces is logically implied, by an induction on deduction length that rests on one lemma: every logical axiom is valid. The only hard case, quantifier instantiation, needs the substitution lemma. The contrapositive corollary states that every satisfiable set is consistent.\n",{"path":17349,"title":17350,"module":17338,"summary":17351},"\u002Flogic\u002Fdeductive-calculus\u002Fcompleteness-and-consistency","The Completeness Theorem","Gödel's completeness theorem is the deep converse of soundness: whatever is logically implied can be deduced. Equivalently, every consistent set has a model. The Henkin proof manufactures that model out of syntax alone: add witnessing constants, extend to a maximal consistent set, and read a term model off the formulas it contains. Compactness and the enumerability theorem drop out.\n",{"path":17353,"title":17354,"module":17355,"summary":17356},"\u002Flogic\u002Fmodels-and-theories\u002Fcompactness-and-lowenheim-skolem","Compactness and the Löwenheim–Skolem Theorems","Models, Compactness, and Theories","A set of first-order sentences has a model whenever each of its finite subsets does. This compactness theorem follows from completeness and yields the finiteness limitation, the downward and upward Löwenheim–Skolem theorems, models of every infinite cardinality, and nonstandard models of arithmetic.\n",{"path":17358,"title":17359,"module":17355,"summary":17360},"\u002Flogic\u002Fmodels-and-theories\u002Ftheories-elementary-classes-and-categoricity","Theories, Elementary Classes, and Categoricity","A theory is a set of sentences closed under logical consequence. Theories correspond to classes of models; a theory may be complete, axiomatizable, or finitely axiomatizable, and completeness together with axiomatizability yields decidability. The Łoś–Vaught test derives completeness from categoricity in a cardinal, applied to dense linear orders and to algebraically closed fields.\n",{"path":17362,"title":17363,"module":17355,"summary":17364},"\u002Flogic\u002Fmodels-and-theories\u002Finterpretations-between-theories","Interpretations Between Theories","An interpretation translates the vocabulary of one theory into formulas of another, relativizing quantifiers to a definable domain and mapping symbols to defining formulas. Defined function symbols meet a noncreativity criterion; the syntactic translation of formulas carries theoremhood forward, and a faithful interpretation transfers decidability and undecidability between theories.\n",{"path":17366,"title":17367,"module":17355,"summary":17368},"\u002Flogic\u002Fmodels-and-theories\u002Fnonstandard-analysis","Nonstandard Analysis","Compactness builds a model of the real ordered field containing infinite elements and nonzero infinitesimals. The transfer principle carries every first-order truth from the reals to this extension, the standard-part map collapses finite hyperreals back onto the reals, and continuity and the derivative are rederived by working with infinitely small quantities directly.\n",{"path":17370,"title":17371,"module":17372,"summary":17373},"\u002Flogic\u002Farithmetic-and-definability\u002Fdefinability-in-arithmetic","The Structure of Arithmetic and Definability","Number Theory and Definability","Number theory is the theory of one fixed structure, the natural numbers under successor, order, addition, multiplication, and exponentiation. Every number is named by a numeral, and a relation is definable when a single formula picks out exactly its tuples. The central gap separates the sentences true in that structure from those any reasonable set of axioms can prove.\n",{"path":17375,"title":17376,"module":17372,"summary":17377},"\u002Flogic\u002Farithmetic-and-definability\u002Fnatural-numbers-with-successor","Natural Numbers with Successor","The weakest reduct keeps only zero and successor. Its models are a standard chain together with disjoint copies of the integers, which makes the theory categorical in every uncountable power, hence complete and decidable. A quantifier-elimination procedure gives a practical decision method and shows a subset is definable if and only if it is finite or cofinite.\n",{"path":17379,"title":17380,"module":17372,"summary":17381},"\u002Flogic\u002Farithmetic-and-definability\u002Fpresburger-and-reducts","Reducts: Order, Addition, and Multiplication","Adding order to the successor reduct keeps decidability and makes the theory finitely axiomatizable; adding addition gives Presburger arithmetic, still decidable by quantifier elimination once congruence predicates are included, with definable sets exactly the eventually periodic ones. Multiplication is the break point: neither addition nor order can define it, and once it joins addition the theory stops being decidable.\n",{"path":17383,"title":17384,"module":17372,"summary":17385},"\u002Flogic\u002Farithmetic-and-definability\u002Fa-subtheory-and-representability","A Subtheory of Number Theory and Representability","A finite set of eleven axioms, the recursion equations for successor, order, addition, multiplication, and exponentiation, already proves every true quantifier-free and existential sentence. Representability asks a theory to prove the right instances of a formula rather than merely make them true, and a relation is defined to be recursive exactly when some consistent finite theory represents it. Church's thesis identifies that with decidability, and closure under composition, minimization, and primitive recursion builds the catalog the incompleteness proofs need.\n",{"path":17387,"title":17388,"module":17389,"summary":17390},"\u002Flogic\u002Fincompleteness\u002Farithmetization-of-syntax","Arithmetization of Syntax","Arithmetization and the Incompleteness Theorems","Gödel numbering assigns a natural number to every symbol, expression, formula, and deduction, turning statements about syntax into statements about numbers. The syntactic operations — substitution, \"is a wff\", \"is an axiom\", \"d codes a deduction of a\" — come out primitive recursive and hence representable in the subtheory, which lets a formula of arithmetic talk about formulas, including itself.\n",{"path":17392,"title":17393,"module":17389,"summary":17394},"\u002Flogic\u002Fincompleteness\u002Fincompleteness-and-undecidability","Incompleteness, Undecidability, and Church's Theorem","The fixed-point lemma manufactures a sentence that talks about its own Gödel number. Pointed at truth it gives Tarski's theorem — arithmetic truth is not arithmetically definable; pointed at provability it gives Gödel's first incompleteness theorem and the undecidability of the theory of the natural numbers, and, applied to validity, Church's theorem that first-order logic is undecidable. The set of theorems of a recursive theory is only recursively enumerable — the gap between provable and true.\n",{"path":17396,"title":17397,"module":17389,"summary":17398},"\u002Flogic\u002Fincompleteness\u002Fsecond-incompleteness-theorem","The Second Incompleteness Theorem","Consistency of a recursively axiomatized theory is itself an arithmetic sentence, built from a provability predicate. When the theory is strong enough to formalize its own reflection and modus ponens — the Hilbert–Bernays–Löb derivability conditions — it cannot prove that sentence unless it is inconsistent. Löb's theorem is the companion result, and set theory is the case that closes Hilbert's program.\n",{"path":17400,"title":17401,"module":17402,"summary":17403},"\u002Flogic\u002Fcomputability-and-representability\u002Frecursive-functions","Recursive Functions and Church's Thesis","Recursive Functions and Representability","The recursive functions are the formal counterpart of the effectively computable ones: built from three initial functions by composition, primitive recursion, and minimization, and equivalently the functions representable in a finitely axiomatized arithmetic. Church's thesis identifies the class with effective calculability; Kleene's normal form theorem and the unsolvable halting problem place the recursive sets strictly inside the recursively enumerable ones.\n",{"path":17405,"title":17406,"module":17402,"summary":17407},"\u002Flogic\u002Fcomputability-and-representability\u002Frepresenting-exponentiation","Representing Exponentiation and the β-Function","Coding finite sequences by prime-power exponents already assumes exponentiation, so representing exponentiation from addition and multiplication alone needs a different encoder. Gödel's β-function, built from a pairing function and the Chinese remainder theorem, reads back arbitrary finite sequences using only plus and times. This represents exponentiation in the addition-multiplication arithmetic and closes the last gap in the representability of every recursive syntactic operation.\n",{"path":17409,"title":17410,"module":17411,"summary":17412},"\u002Flogic\u002Fsecond-order-logic\u002Fsecond-order-languages","Second-Order Languages","Second-Order Logic and Beyond","Second-order logic quantifies over relations and functions, not just individuals. Second-order Peano arithmetic and the second-order theory of the reals become categorical, and finiteness is definable by a single sentence. Compactness, completeness, and the Löwenheim–Skolem theorems all fail for the standard semantics.\n",{"path":17414,"title":17415,"module":17411,"summary":17416},"\u002Flogic\u002Fsecond-order-logic\u002Fskolem-functions-and-many-sorted-logic","Skolem Functions and Many-Sorted Logic","Skolem functions replace existential quantifiers with named witnesses, putting any first-order formula into a prenex form with all existentials — now over functions — pulled to the front. The Skolemized formula is equisatisfiable with the original, which reduces satisfiability to universal sentences and, through Herbrand expansions, to sentential logic. Many-sorted logic then adds several universes at once and reduces cleanly to ordinary one-sorted logic.\n",{"path":17418,"title":17419,"module":17411,"summary":17420},"\u002Flogic\u002Fsecond-order-logic\u002Fgeneral-structures","General (Henkin) Structures","General semantics reinterprets second-order logic by letting the predicate and function quantifiers range over a designated collection of relations and functions rather than all of them. Recast as many-sorted first-order logic with comprehension axioms, general second-order logic recovers a sound and complete calculus together with compactness and Löwenheim–Skolem, giving up the categoricity of the standard semantics. The ω-models of analysis show the trade.\n",{"path":17422,"title":17423,"module":6,"summary":6},"\u002Flogic","Logic",{"path":17425,"title":17426,"module":14054,"summary":17427},"\u002Freinforcement-learning\u002Ffoundations\u002Fwhat-is-reinforcement-learning","What Is Reinforcement Learning?","Reinforcement learning is learning what to do — how to map situations to actions — so as to maximize a numerical reward signal, discovered by trial and error rather than told. We set up the agent–environment loop, separate it from supervised and unsupervised learning, name the four elements (policy, reward, value, and an optional model), and train a tic-tac-toe player with a temporal-difference value update.\n",{"path":17429,"title":17430,"module":14054,"summary":17431},"\u002Freinforcement-learning\u002Ffoundations\u002Fa-brief-history-of-rl","A Brief History of Reinforcement Learning","The origins of reinforcement learning. Three threads — trial-and-error learning from animal psychology, optimal control and dynamic programming, and temporal-difference learning — ran independently for decades and merged around 1989 into the modern field. Replacing the lookup table with a neural network then produced deep reinforcement learning: DQN, AlphaGo, AlphaZero, MuZero, and RLHF.\n",{"path":17433,"title":17434,"module":14054,"summary":17435},"\u002Freinforcement-learning\u002Ffoundations\u002Fmulti-armed-bandits","Multi-Armed Bandits","A bandit is reinforcement learning stripped to a single decision, repeated: no state, no consequences, only the tension between exploiting the arm that looks best and exploring the ones that might be better. We build up the whole toolkit — sample-average value estimates, the incremental update rule, ε-greedy, optimistic initialization, UCB, and gradient bandits — and use it to study exploration in isolation, the one problem that carries over to the full setting.\n",{"path":17437,"title":17438,"module":14054,"summary":17439},"\u002Freinforcement-learning\u002Ffoundations\u002Fbandit-exploration-algorithms","Bandit Exploration Algorithms","Better ways to explore than picking at random. Upper-confidence-bound selection explores by optimism about what it hasn't measured; gradient bandits learn action preferences by stochastic gradient ascent on reward. We then add context to get the contextual bandit, the bridge to full RL, and measure everything by regret — where UCB1 and Thompson sampling reach the logarithmic optimum that fixed-ε greedy cannot.\n",{"path":17441,"title":17442,"module":14054,"summary":17443},"\u002Freinforcement-learning\u002Ffoundations\u002Fmarkov-decision-processes","Markov Decision Processes","A Markov decision process is the formal interface between an agent and its environment: at each step the agent reads a state, chooses an action, and receives a reward and a next state. We fix that loop, the dynamics function that governs it, and the Markov property that makes the state sufficient; then turn goals into a scalar reward and rewards into a discounted return, with one notation that covers both episodic and continuing tasks.\n",{"path":17445,"title":17446,"module":14054,"summary":17447},"\u002Freinforcement-learning\u002Ffoundations\u002Fvalue-functions-and-optimality","Value Functions and Optimality","A value function scores how good a state (or state–action pair) is under a policy: the expected return from there onward. Its defining property is the Bellman equation, a self-consistency condition linking a state's value to its successors' values, which we derive from the return and the dynamics. Pushing the same idea to the best-achievable value gives the Bellman optimality equations, whose solution yields an optimal policy — and whose intractability is what the rest of the course is about.\n",{"path":17449,"title":14273,"module":17450,"summary":17451},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fdynamic-programming","Tabular Solution Methods","Dynamic programming computes optimal policies when a perfect model of the MDP is given, by turning the Bellman equations into assignment statements. We build up iterative policy evaluation (the expected update), the policy improvement theorem, and the two classic algorithms that alternate them — policy iteration and value iteration — worked on the gridworld, a two-state MDP, Jack's car rental, and the gambler's problem.\n",{"path":17453,"title":17454,"module":17450,"summary":17455},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fdp-async-and-gpi","Dynamic Programming: Asynchronous DP and Generalized Policy Iteration","Policy and value iteration both sweep the entire state set on every pass, which is impossible once the state space is huge. This lesson loosens the schedule: asynchronous DP updates states in any order, generalized policy iteration names the alternation of evaluation and improvement that underlies nearly every RL method, and a look at efficiency and the curse of dimensionality places DP among the alternatives. We close past Sutton & Barto with prioritized sweeping, neuro-dynamic programming, value-iteration networks, and MuZero.\n",{"path":17457,"title":17458,"module":17450,"summary":17459},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fmonte-carlo-methods","Monte Carlo Methods","Monte Carlo methods learn value functions and optimal policies from complete sampled episodes, with no model of the environment: they simply average the returns that actually followed each state. We build prediction (first-visit and every-visit averaging), see why estimating action values forces the exploration question, and answer it two ways on-policy — exploring starts and epsilon-soft control. Throughout, Monte Carlo samples one whole trajectory to termination and never bootstraps.\n",{"path":17461,"title":17462,"module":17450,"summary":17463},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fmonte-carlo-off-policy","Monte Carlo Methods: Off-Policy Learning","On-policy Monte Carlo can only reach the best exploring policy, not the true optimum. Off-policy methods remove that ceiling by learning about a greedy target policy from data generated by a soft behavior policy, corrected with importance sampling. We derive the importance-sampling ratio, weigh ordinary against weighted estimators on real numbers, give the incremental off-policy algorithm, sharpen it with discounting-aware sampling, and close by placing Monte Carlo on the model\u002Fbootstrap map beside DP and temporal-difference learning.\n",{"path":17465,"title":17466,"module":17450,"summary":17467},"\u002Freinforcement-learning\u002Ftabular-methods\u002Ftemporal-difference-learning","Temporal-Difference Learning","Temporal-difference learning is the one idea most central to reinforcement learning: learn a value directly from experience, like Monte Carlo, but update each guess toward the next guess before the episode ends, like dynamic programming. We derive the TD(0) prediction rule and its reward-prediction error, contrast its one-step backup with MC and DP, work the driving-home and random-walk examples, and show the batch-updating optimality that makes TD approximate the certainty-equivalence estimate.\n",{"path":17469,"title":17470,"module":17450,"summary":17471},"\u002Freinforcement-learning\u002Ftabular-methods\u002Ftd-control-sarsa-and-q-learning","TD Control: Sarsa, Q-learning, and Double Learning","With TD prediction in hand, control follows the generalized-policy-iteration pattern with TD as the evaluation step. We build Sarsa (on-policy), Q-learning (off-policy, targeting the optimal policy), and Expected Sarsa that spans the two, then confront the maximization bias every max-based method inherits and fix it with Double Q-learning. We close past Sutton & Barto, following each one-step tabular update into its deep-RL descendant — DQN, Double DQN, and Rainbow.\n",{"path":17473,"title":17474,"module":17450,"summary":17475},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fn-step-bootstrapping","n-Step Bootstrapping","Monte Carlo waits for the full return; one-step TD bootstraps after a single reward. Between them lies a whole spectrum, indexed by one integer n: look ahead n real rewards, then bootstrap from the value n steps out. The n-step return unifies the previous two lessons, and — on the random walk — an intermediate n beats both extremes. We build the n-step return, the n-step TD update, the backup-diagram spectrum, and n-step Sarsa for control.\n",{"path":17477,"title":17478,"module":17450,"summary":17479},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fn-step-off-policy-methods","n-Step Bootstrapping: Off-Policy Methods","Taking the n-step family off-policy raises the same importance-sampling questions Monte Carlo did, now over a window of exactly n actions. We reweight n-step returns by the policy ratio, watch the ratio product inflate variance on real numbers, then build the tree-backup algorithm that learns off-policy with no ratios at all — and finally n-step Q(sigma), one algorithm whose per-step switch recovers Sarsa, tree backup, and Expected Sarsa as special cases.\n",{"path":17481,"title":17482,"module":17450,"summary":17483},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fplanning-and-learning","Planning and Learning","Planning and learning are the same operation run on two kinds of experience. A model turns states and actions into simulated transitions; planning backs up values over that simulated experience exactly as learning backs them up over real experience. We build the Dyna architecture that interleaves acting, model-learning, direct RL, and planning in one loop, trace a single Dyna-Q step by hand, and patch the architecture for when the model goes stale.\n",{"path":17485,"title":17486,"module":17450,"summary":17487},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fplanning-focusing-and-decision-time","Planning: Focusing Updates and Decision-Time Search","Dyna plans by replaying remembered transitions, but sampling them uniformly wastes most of the effort. This lesson sharpens planning: prioritized sweeping works backward from states whose value just changed, expected versus sample updates weigh thoroughness against cost, and trajectory sampling and real-time DP focus updates on the states the policy actually visits. We trace Dyna forward to model-based deep RL, then turn to decision-time planning — heuristic search, rollouts, and Monte Carlo Tree Search.\n",{"path":17489,"title":17490,"module":17450,"summary":17491},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fdecision-time-planning","Decision-Time Planning","Planning need not build a global policy. Decision-time planning runs a fresh lookahead every time a state arrives and returns just one action, then throws the work away. We start from real-time dynamic programming — asynchronous value iteration on the states the agent actually visits — then move through heuristic search and rollout algorithms, each a one-step policy improvement applied on the fly to the current state.\n",{"path":17493,"title":17494,"module":17450,"summary":17495},"\u002Freinforcement-learning\u002Ftabular-methods\u002Fmonte-carlo-tree-search","Monte Carlo Tree Search","Monte Carlo Tree Search is a rollout algorithm with memory: it accumulates value estimates across simulations and steers later ones toward promising branches. We work through the four steps — selection, expansion, simulation, backup — the UCT selection rule computed on real numbers, the asymmetric growing tree, and the full pseudocode. We close past Sutton & Barto with the lineage from UCT to AlphaGo, AlphaZero, and MuZero, where a learned network stands in for the leaf value and the rollout.\n",{"path":17497,"title":17498,"module":17499,"summary":17500},"\u002Freinforcement-learning\u002Fapproximation\u002Fon-policy-prediction","On-Policy Prediction with Approximation","Approximate Solution Methods","Every tabular method so far stored one number per state, which fails once the state space is large or continuous. We replace the table with a parameterized value function $\\hat v(s,\\mathbf{w})$, define the mean squared value error it should minimize under the on-policy distribution, and derive stochastic- and semi-gradient learning rules — the semi-gradient TD(0) update that bootstraps and so is not a true gradient. Linear methods make the analysis clean and give the TD fixed point; feature construction (polynomials, Fourier basis, coarse and tile coding, RBFs) supplies the vectors $\\mathbf{x}(s)$, and neural networks are the nonlinear bridge to deep RL.\n",{"path":17502,"title":17503,"module":17499,"summary":17504},"\u002Freinforcement-learning\u002Fapproximation\u002Ffeature-construction-and-nonlinear","Feature Construction and Nonlinear Approximation","Linear methods are only as good as the feature vectors $\\mathbf{x}(s)$ fed to them, and this lesson builds those vectors. Polynomials and the Fourier basis turn a state's coordinates into smooth global features; coarse coding, tile coding, and radial basis functions cover a continuous space with overlapping local receptive fields whose size sets the reach of generalization. Then we stop designing features by hand: a neural network learns the representation itself by gradient descent, trading the convergence guarantees of the linear case for expressiveness — the bridge to deep reinforcement learning.\n",{"path":17506,"title":17507,"module":17499,"summary":17508},"\u002Freinforcement-learning\u002Fapproximation\u002Fon-policy-control","On-Policy Control with Approximation","Prediction learned a value function from features; control learns to act. We carry semi-gradient methods over to action values $\\hat q(s,a,\\mathbf{w})$, giving episodic semi-gradient Sarsa and its n-step form, and solve Mountain Car by descending a cost-to-go surface. In the continuing case, function approximation makes discounting unable to affect which policy is best, so we replace it with the average-reward setting — the differential return, differential value functions, and differential semi-gradient Sarsa.\n",{"path":17510,"title":17511,"module":17499,"summary":17512},"\u002Freinforcement-learning\u002Fapproximation\u002Faverage-reward-control","Average-Reward Control for Continuing Tasks","With function approximation, discounting has no effect on a continuing task: averaged over the on-policy distribution, the discounted objective equals the average reward times a policy-independent constant, so $\\gamma$ cannot change which policy is best. This lesson replaces discounting with the average-reward setting — the long-run reward rate $r(\\pi)$, the differential return that measures each state's transient advantage over that rate, differential value functions and TD error, and differential semi-gradient Sarsa, the control method for continuing tasks that never invokes a discount factor.\n",{"path":17514,"title":17515,"module":17499,"summary":17516},"\u002Freinforcement-learning\u002Fapproximation\u002Foff-policy-and-the-deadly-triad","Off-Policy Methods and the Deadly Triad","Off-policy learning with function approximation is where the convergence guarantees of reinforcement learning fail. We extend the tabular off-policy updates to semi-gradient form with per-step importance sampling, show Baird's counterexample driving the weights to infinity, and identify the cause: the deadly triad of function approximation, bootstrapping, and off-policy training — any two are safe, all three can diverge. The divergence is not caused by sampling noise: a fully synchronous dynamic-programming update blows up just the same, which is what makes the triad a structural hazard rather than a fluke.\n",{"path":17518,"title":17519,"module":17499,"summary":17520},"\u002Freinforcement-learning\u002Fapproximation\u002Fbellman-error-and-gradient-td","Value-Function Geometry and Gradient-TD Methods","Why does the deadly triad diverge, and how do you stop it? This lesson develops the geometry that explains the failure: value functions as vectors, the projection operator onto the representable subspace, and the split between the Bellman error, the value error, and the projected Bellman error: the three objectives have different minimizers. The projected Bellman error is the learnable one, and Gradient-TD methods (GTD2, TDC) do true stochastic gradient descent on it, staying stable even off-policy at $O(d)$ cost. Emphatic TD reweights states instead, and a survey of variance-reduction techniques closes the gap between stability and usable learning.\n",{"path":17522,"title":17523,"module":17499,"summary":17524},"\u002Freinforcement-learning\u002Fapproximation\u002Feligibility-traces","Eligibility Traces","n-step methods unify TD and Monte Carlo by storing the last n feature vectors; eligibility traces do the same job with a single short-term memory vector. The λ-return averages every n-step return under a geometric weighting; the forward view looks ahead to that average, and the backward view produces nearly the same updates online through a decaying trace vector. We build the λ-return, TD(λ) with its trace, the two ways λ recovers TD(0) and Monte Carlo, a note on the exact equivalence of true online TD(λ), and Sarsa(λ) for control.\n",{"path":17526,"title":17527,"module":17499,"summary":17528},"\u002Freinforcement-learning\u002Fapproximation\u002Ftrue-online-and-sarsa-lambda","True Online TD(λ) and Sarsa(λ)","Plain TD(λ) makes the forward and backward views nearly agree; this lesson closes the gap. True online TD(λ) uses a dutch trace and a small correction term to produce exactly the same weight sequence as the online λ-return algorithm, at the same memory and only a constant factor more compute — the sharpest statement of the forward\u002Fbackward duality. The whole apparatus then lifts to control unchanged: Sarsa(λ) threads a single delayed reward back along an entire trajectory in one sweep, and the λ-weighting reappears in modern deep RL as generalized advantage estimation.\n",{"path":17530,"title":17531,"module":17499,"summary":17532},"\u002Freinforcement-learning\u002Fapproximation\u002Fpolicy-gradient-methods","Policy Gradient Methods","Every method so far learned values and read a policy off them. Policy gradient methods drop the intermediary: parameterize the policy directly and climb the performance gradient. We build the softmax-in-preferences parameterization, prove the policy gradient theorem that makes the gradient computable without the unknown state distribution, and derive REINFORCE and its variance-cutting state-value baseline — the launch point for the bootstrapping actor-critic that follows.\n",{"path":17534,"title":17535,"module":17499,"summary":17536},"\u002Freinforcement-learning\u002Fapproximation\u002Factor-critic-and-continuous-actions","Actor-Critic Methods and Continuous Actions","REINFORCE with a baseline learns a value function but never bootstraps; this lesson adds the bootstrapping critic that completes the actor-critic architecture. The critic scores each transition into a single TD error that steers both the actor's policy step and its own value step, trading a little bias for much lower variance and fully online, continuing-task learning. The policy gradient theorem carries over unchanged to the average-reward setting, a Gaussian policy handles real-valued actions with self-tuning exploration, and the natural policy gradient leads straight to TRPO, PPO, and the deep actor-critic methods that train today's agents.\n",{"path":17538,"title":17539,"module":17499,"summary":17540},"\u002Freinforcement-learning\u002Fapproximation\u002Fleast-squares-and-memory-based-methods","Least-Squares TD","Semi-gradient TD spends one cheap step per example and needs many examples; this lesson makes the opposite tradeoff. Least-Squares TD (LSTD) accumulates the matrices $\\mathbf{A}$ and $\\mathbf{b}$ and solves the TD fixed point $\\mathbf{w} = \\mathbf{A}^{-1}\\mathbf{b}$ directly, using the Sherman-Morrison identity to maintain the inverse in $O(d^2)$ — the most data-efficient linear TD method, at a quadratic cost. We work a solve by hand, weigh the quadratic cost against semi-gradient TD's cheap steps, and note that LSTD never forgets — a problem in control, where least-squares policy iteration is the natural extension.\n",{"path":17542,"title":17543,"module":17499,"summary":17544},"\u002Freinforcement-learning\u002Fapproximation\u002Fmemory-and-kernel-methods","Memory-Based and Kernel Methods","Least-squares TD spent more compute to extract more from each example; this lesson drops the parametric form entirely. Memory-based methods store training examples untouched and answer a query locally at retrieval time — nearest neighbor, weighted average, locally weighted regression — so accuracy grows with the data and effort concentrates where the agent actually goes. Kernel-based methods weight stored examples by a similarity kernel $k(s,s')$, and every linear method turns out to be a kernel method. Interest and emphasis, finally, make the on-policy weighting itself a design choice, aiming scarce approximation capacity at the states that matter.\n",{"path":17546,"title":17547,"module":17499,"summary":17548},"\u002Freinforcement-learning\u002Fapproximation\u002Foff-policy-eligibility-traces","Off-Policy Eligibility Traces","Eligibility traces meet off-policy learning and function approximation — the corner where stability gets hard. We first let the bootstrapping and discounting parameters vary with state, so a single generalized return covers episodic and continuing tasks and folds termination into the discount. Then we fold the per-decision importance ratio into the trace with a control-variate correction, and build Watkins's Q(λ) and its importance-sampling-free successor Tree-Backup(λ) — all correct in expectation, but still semi-gradient, so the deadly triad and its fixes wait for the next lesson.\n",{"path":17550,"title":17551,"module":17499,"summary":17552},"\u002Freinforcement-learning\u002Fapproximation\u002Fstable-off-policy-traces","Stable Off-Policy Methods with Traces","Off-policy traces get the expected target right, but with $\\lambda \u003C 1$ they bootstrap, so off-policy plus bootstrapping plus function approximation is the deadly triad and the weights can diverge. This lesson carries the two one-step fixes to traces: GTD(λ) and GQ(λ) add a second weight vector and a gradient correction for true gradient descent on the projected Bellman error, while Emphatic TD(λ) reweights updates through a followon trace and interest to recover the on-policy stability. It closes with the implementation reality that traces are cheap because they are sparse, and with Retrace and V-trace — the clipped-ratio descendants that make off-policy traces work at deep-RL scale.\n",{"path":17554,"title":16872,"module":17555,"summary":17556},"\u002Freinforcement-learning\u002Fdeep-rl\u002Fdeep-q-networks","Deep Reinforcement Learning","Deep Q-networks replace the linear value function with a neural network $Q(s,a;\\mathbf{w})$ and confront the fact that a nonlinear approximator, off-policy bootstrapping, and correlated online data — the deadly triad — make naive Q-learning diverge. DQN counters this empirically with two stabilizers: an experience replay buffer that decorrelates and reuses samples, and a periodically-frozen target network that fixes the bootstrap target. We derive the DQN loss and gradient, walk through the Atari convolutional architecture and its results, and then add the three refinements that define modern value-based deep RL — Double DQN, dueling networks, and prioritized experience replay.\n",{"path":17558,"title":17559,"module":17555,"summary":17560},"\u002Freinforcement-learning\u002Fdeep-rl\u002Fdqn-improvements","DQN Improvements: Double, Dueling, and Prioritized Replay","Three refinements that turn plain DQN into the standard modern value-based agent, each touching a different part of the system. Double DQN fixes the maximization bias in the target by splitting action selection from evaluation; dueling networks restructure the network around a state value and per-action advantages; prioritized replay changes which transitions are learned from. We close with Rainbow, which combines them, and the distributional view that predicts the whole return distribution rather than its mean.\n",{"path":17562,"title":17563,"module":17555,"summary":17564},"\u002Freinforcement-learning\u002Fdeep-rl\u002Factor-critic-and-ppo","Actor–Critic and GAE","Make the actor and the critic deep networks and the policy-gradient architecture becomes modern deep RL. We build the neural actor-critic, the advantage estimate that replaces the raw return, and Generalized Advantage Estimation as a λ-blend of n-step advantages, then the parallel-worker methods A3C and A2C that decorrelate on-policy data. The step-size constraints — trust regions, PPO, and the continuous-control family — follow in the next lesson.\n",{"path":17566,"title":17567,"module":17555,"summary":17568},"\u002Freinforcement-learning\u002Fdeep-rl\u002Fppo-and-continuous-control","PPO and Continuous Control","Keeping the policy-gradient step from destroying the policy, and the algorithms that result. Trust-region optimization bounds each update by a KL constraint; PPO keeps that goal but replaces the second-order machinery with a first-order clip on the probability ratio, which is why it is the modern default and the optimizer inside RLHF. We then tour the off-policy continuous-control family — DDPG, TD3, and SAC — and where actor-critic went at scale, from OpenAI Five to language-model alignment.\n",{"path":17570,"title":17571,"module":17555,"summary":17572},"\u002Freinforcement-learning\u002Fdeep-rl\u002Fcase-studies","Case Studies: Learning to Play","The game-playing systems that turned reinforcement learning from a theory into a track record: Samuel's checkers player, TD-Gammon, Watson's Daily-Double wagering, a reinforcement-learning memory controller, DQN, and AlphaGo through AlphaGo Zero. Read as a set they draw one line — a value function, learned by self-play or interaction, refined by search, carried by a deep network — that runs from a 1959 checkers program to superhuman Go.\n",{"path":17574,"title":17575,"module":17555,"summary":17576},"\u002Freinforcement-learning\u002Fdeep-rl\u002Frl-beyond-games","Reinforcement Learning Beyond Games","The same value-and-reward machinery, pointed at problems with no opponent. Web personalization as a contextual bandit and then a full MDP for life-time value; thermal soaring, where a glider learns to climb on turbulent air and reward design does most of the work; and the industrial-scale systems that carried the same design past Sutton & Barto — AlphaStar, OpenAI Five, GT Sophy, and RLHF, where the reward itself is learned from human preference.\n",{"path":17578,"title":17579,"module":17555,"summary":17580},"\u002Freinforcement-learning\u002Fdeep-rl\u002Ffrontiers","Frontiers: Beyond the Standard MDP","The standard MDP fixes three things — state, reward, and single-step actions — and this lesson loosens two of them. We generalize the value function into a general value function that predicts any signal, and use those predictions as auxiliary tasks that shape representations; we extend actions in time with the options framework; and we treat state as a construction the agent builds from a stream of observations. Reward design and the open problems follow in the next lesson.\n",{"path":17582,"title":17583,"module":17555,"summary":17584},"\u002Freinforcement-learning\u002Fdeep-rl\u002Freward-design-and-open-problems","Reward Design and Open Problems","How to design a reward signal that encodes the intended goal — sparse reward, shaping, and reward hacking — and the problems the whole tabular, approximate, and deep arc leaves unsolved. We close with how the frontiers were pushed after Sutton & Barto: auxiliary tasks, learned options, intrinsic-motivation bonuses, learned world models, and offline RL, then the two concerns of reward hacking and safety that any real-world agent must address.\n",{"path":17586,"title":17587,"module":17588,"summary":17589},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fdistributional-and-rainbow","Sharpening DQN: Improvements and the Distributional Idea","Modern Deep Reinforcement Learning","In the years after the 2015 DQN paper, a stream of focused improvements each fixed one weakness of the baseline without disturbing its frame. This lesson recaps five that keep the scalar $Q$-value — Double DQN, multi-step returns, dueling networks, prioritized replay, and NoisyNets, each changing a different slot of the same Q-learning loop — then develops the sixth, distributional RL, which changes the objective itself: learn the whole return distribution $Z(s,a)$. We build the distributional Bellman equation and the C51 categorical algorithm, projection step and all, worked end to end on real numbers. A companion lesson takes up QR-DQN, Rainbow, and the modern distributional line.\n",{"path":17591,"title":17592,"module":17588,"summary":17593},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fdistributional-and-rainbow-part-2","Distributional RL and Rainbow","A companion to the DQN improvements lesson. C51 fixed the return atoms and learned their probabilities; QR-DQN does the reverse — fix the probabilities, learn the values — which removes the projection and trains with a quantile loss. We cover why the distribution helps even when you act on the mean, then assemble Rainbow: all six improvements in one Q-learning loop, with the component ablation that shows each one's real weight. The distributional line then runs on through IQN, FQF, and Agent57, the first agent to beat the human baseline on all 57 Atari games.\n",{"path":17595,"title":17596,"module":17588,"summary":17597},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fcontinuous-control","Continuous Control: DDPG and TD3","When actions are real-valued, the $\\arg\\max_a Q(s,a)$ in Q-learning becomes an optimization problem on every step. This lesson builds the off-policy actor-critic family that sidesteps it: the deterministic policy gradient and DDPG, which replaces the max with a learned actor, and the three fixes of TD3 that counter the value overestimation DDPG inherits. A companion lesson takes up SAC's maximum-entropy objective and the methods built on this template.\n",{"path":17599,"title":17600,"module":17588,"summary":17601},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fcontinuous-control-part-2","Continuous Control: SAC and Beyond","A companion to the DDPG and TD3 lesson. Where those actors are deterministic and explore with bolted-on noise, soft actor-critic (SAC) changes the objective itself: maximize return plus the entropy of the policy, so exploration becomes intrinsic and the agent stays robust. We develop the maximum-entropy objective, the reparameterized squashed-Gaussian actor, and automatic temperature tuning, then survey the methods built on this off-policy template — distributional critics (D4PG), critic ensembles (REDQ), and control from pixels (DrQ, RAD).\n",{"path":17603,"title":17604,"module":17588,"summary":17605},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmodel-based-rl","Model-Based Deep RL: Sample Efficiency and PETS","A model turns experience into imagined planning. This lesson makes the sample-efficiency case for learning a dynamics model, works through why a learned model's errors compound over the planning horizon, and builds the most direct model-based method: PETS plans online with a probabilistic ensemble under model-predictive control, distrusting the model exactly where its members disagree. A companion lesson takes up latent world models (Dreamer) and MuZero.\n",{"path":17607,"title":17608,"module":17588,"summary":17609},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmodel-based-rl-part-2","Model-Based Deep RL: World Models, Dreamer, and MuZero","A companion to the PETS lesson. PETS plans in the environment's native state space; these methods change what the model represents. World Models and Dreamer learn a compact latent state and do almost all their learning by imagining inside it, with value gradients flowing through the differentiable dynamics. MuZero predicts neither states nor pixels — only the reward, value, and policy that MCTS reads — and plans with search against that learned model, AlphaZero without the rules. We close with MBPO, TD-MPC, and EfficientZero.\n",{"path":17611,"title":17612,"module":17588,"summary":17613},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fexploration","Exploration in Deep RL: Novelty as Reward","When the state space is enormous and reward is rare, ε-greedy amounts to a random walk that almost never reaches the first reward. This lesson scales the bandit's exploration ideas up to deep RL through the dominant approach — manufacture a reward for novelty and let the agent chase it: optimism and pseudo-counts from density models, and intrinsic motivation and curiosity (the Intrinsic Curiosity Module and Random Network Distillation). A companion lesson takes up posterior sampling, Go-Explore, and the modern methods.\n",{"path":17615,"title":17616,"module":17588,"summary":17617},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fexploration-part-2","Exploration in Deep RL: Posterior Sampling and Go-Explore","A companion to the novelty-as-reward lesson. Pseudo-counts and curiosity reward the unfamiliar after the agent stumbles into it; this lesson covers two ideas that go further. Bootstrapped DQN keeps an ensemble that approximates a posterior over value functions and explores by committing to one sampled hypothesis per episode — the deep, directed exploration ε-greedy cannot manage. Go-Explore remembers and returns to the frontier, defeating detachment and derailment to solve Montezuma's Revenge. We close with episodic memory (Never Give Up), Agent57, and model-based exploration.\n",{"path":17619,"title":17620,"module":17588,"summary":17621},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Foffline-rl","Offline RL: The Problem and Value-Based Fixes","Offline reinforcement learning learns a policy from a fixed logged dataset with no further environment interaction — off-policy learning pushed to the extreme, and it breaks for the extreme version of the same reason. Bootstrapping queries the value function at out-of-distribution actions the data never covers, those errors are optimistic, and with no online feedback to correct them they compound through the Bellman backup. This lesson sets up the failure and off-policy evaluation, then builds the first two families of pessimistic fixes: policy constraint (BCQ) and conservative value estimation (CQL). A companion lesson takes up implicit methods, model-based offline RL, and Decision Transformer.\n",{"path":17623,"title":17624,"module":17588,"summary":17625},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Foffline-rl-part-2","Offline RL: Implicit Methods, Sequence Models, and Beyond","A companion to the offline-RL problem lesson. Policy constraint and conservative value estimation both still query a learned value function; implicit methods (IQL) avoid querying it off the data at all, using an in-sample expectile backup. We then build pessimism into a learned model (MOPO, COMBO) and drop bootstrapping entirely with Decision Transformer's return-conditioned sequence modeling, closing with offline-to-online fine-tuning, diffusion planners, and the offline view of RLHF. The one rule throughout: without online correction, be pessimistic about what you cannot verify.\n",{"path":17627,"title":17628,"module":17588,"summary":17629},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fimitation-and-inverse-rl","Imitation Learning: Cloning, DAgger, and Inverse RL","When a reward is hard to specify but an expert is easy to watch, learn from demonstrations instead. Behavioral cloning treats control as supervised learning of the expert's state-to-action map, and fails through compounding error: small mistakes carry the agent off the expert's distribution, where it was never trained. DAgger fixes the mismatch by querying the expert on the learner's own states. Inverse RL instead recovers the reward the expert seems to optimize — an ill-posed problem that maximum-entropy IRL disambiguates. A companion lesson casts imitation as adversarial occupancy matching (GAIL, AIRL).\n",{"path":17631,"title":17632,"module":17588,"summary":17633},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fimitation-and-inverse-rl-part-2","Imitation as Adversarial Matching: GAIL and AIRL","A companion to the imitation-learning lesson. If the point of recovering a reward is only to re-run RL and match the expert, you can skip the reward and match the behavior directly. GAIL casts imitation as a GAN — a discriminator separating expert from learner state-action pairs supplies the reward a policy-gradient method optimizes — matching occupancy measures without ever naming a reward. AIRL reads a transferable reward back out of the discriminator. We compare all four methods and close with reward models in RLHF, scaled cloning, and diffusion policies.\n",{"path":17635,"title":17636,"module":17588,"summary":17637},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmulti-agent-rl","Multi-Agent RL: Markov Games and Centralized Training","With more than one learning agent in an environment, each agent's world becomes non-stationary because the others are changing too. This lesson builds the Markov-game generalization of the MDP, diagnoses non-stationarity as the central obstacle, shows why the naive baselines fail, and develops the dominant fix — centralized training with decentralized execution (MADDPG, VDN, QMIX). A companion lesson takes up self-play, the landmark game-playing systems, and the equilibrium concepts that define what \"solved\" means.\n",{"path":17639,"title":17640,"module":17588,"summary":17641},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmulti-agent-rl-part-2","Multi-Agent RL: Self-Play and Solution Concepts","A companion to the Markov-games lesson. In the purely competitive setting, an agent can generate its own training curriculum by playing against copies of itself — self-play, the method behind AlphaGo, OpenAI Five, and AlphaStar. We develop why self-play produces an ever-improving opponent, the systems it built, and then the equilibrium solution concepts (Nash, correlated, coarse-correlated) that define what \"solved\" means once there is an opponent, closing with PSRO, MAPPO, and the language-model-agent frontier.\n",{"path":17643,"title":17644,"module":17588,"summary":17645},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fhierarchical-rl","Hierarchical RL: Options and the Option-Critic","Flat RL cannot explore a long horizon: reaching reward through hundreds of primitive actions is exponentially unlikely, and every credit-assignment update crawls one step at a time. Hierarchy breaks one hard long-horizon problem into many short ones. This lesson develops temporal abstraction — the options framework and its semi-Markov view, and learning options end to end with the option-critic. A companion lesson takes up goal-conditioned manager\u002Fworker hierarchies (FeUdal Networks and HIRO), hindsight relabeling, and unsupervised skill discovery.\n",{"path":17647,"title":17648,"module":17588,"summary":17649},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fhierarchical-rl-part-2","Hierarchical RL: Goal-Conditioned Hierarchies and Skills","A companion to the options lesson. Options package a behavior; goal-conditioned hierarchies instead give the top level an explicit language of goals — a manager proposes a target state or a latent direction, and a worker is rewarded for reaching it (FeUdal Networks, HIRO). We develop that architecture, the hindsight relabeling that lets it learn from sparse reward, and unsupervised skill discovery (DIAYN) that learns a repertoire of behaviors with no reward at all. The shared idea throughout: shorten the horizon by inserting a level that decides less often.\n",{"path":17651,"title":17652,"module":17588,"summary":17653},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Frlhf-and-language-models","RLHF and Language Models","A language model trained to predict the next token is fluent but not helpful, honest, or harmless — the objective it was optimized for is not the objective we want. RLHF closes that gap by turning the one thing humans do reliably, comparing two outputs, into a reward. We build the three-stage pipeline: supervised fine-tuning, a Bradley-Terry reward model fit to preference pairs, then PPO against that reward with a KL penalty keeping it near the reference policy. We then cover reward hacking and why the KL penalty matters, Direct Preference Optimization, which folds the reward model into a single classification loss, and the RLAIF and verifiable-reward variants. This pipeline is what makes the largest models usable as assistants.\n",{"path":17655,"title":17656,"module":17588,"summary":17657},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fpartial-observability-pomdps","Partial Observability: POMDPs and the Belief State","Drop the assumption that the agent sees the state. It sees an observation, a partial and noisy function of a hidden state, and one observation is no longer a Markov signal. This lesson builds the POMDP tuple, shows that the belief state — the posterior over hidden states — is a sufficient statistic that turns a POMDP back into an MDP over beliefs, and works the Bayes-filter belief update step by step. A companion lesson explains why exact planning is intractable and develops the deep-RL answer of recurrent, history-based policies.\n",{"path":17659,"title":17660,"module":17588,"summary":17661},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fpartial-observability-pomdps-part-2","Partial Observability: Planning and Recurrent Policies","A companion to the belief-state lesson. In principle a POMDP reduces to an MDP over beliefs; in practice two obstacles block that. Exact planning over the belief simplex is intractable — the value function is piecewise-linear-and-convex with a number of pieces that can explode — and computing the belief needs a model the agent rarely has. This lesson develops the intractability, the point-based approximations that address it, and the deep-RL answer: make the policy a function of history with a recurrent network (DRQN, R2D2), with frame-stacking, attention, and world-model latents as learned beliefs.\n",{"path":17663,"title":17664,"module":17588,"summary":17665},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fsafe-and-constrained-rl","Safe and Constrained RL: The CMDP and Policy Methods","Maximizing a scalar reward is not the same as behaving well: a capable optimizer will find and exploit any gap between the reward and what its designer actually meant, a failure called specification gaming or reward hacking. The remedy is to add explicit cost constraints — the constrained MDP — maximizing return subject to an expected-cost budget. This lesson builds the core toolkit: the CMDP itself, Lagrangian primal-dual methods that learn a multiplier on the constraint (RCPO), and constrained policy optimization (CPO) with its trust-region cost bound. A companion lesson covers risk-sensitivity, safe exploration, and the alignment framing.\n",{"path":17667,"title":17668,"module":17588,"summary":17669},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fsafe-and-constrained-rl-part-2","Safe RL: Risk, Safe Exploration, and Alignment","A companion to the constrained-MDP lesson. Constraining the mean cost is not enough: a policy safe on average can be catastrophic in the tail, and a policy safe at convergence can violate its limits wildly while learning. This lesson optimizes the tail with risk-sensitive objectives (CVaR), then makes exploration itself safe with shields, Lyapunov methods, and safety layers that project unsafe actions onto the feasible set — closing with benchmarks, safe RLHF, robustness, and the alignment framing that ties safety back to the problem of incompletely specified reward.\n",{"path":17671,"title":17672,"module":17588,"summary":17673},"\u002Freinforcement-learning\u002Fmodern-deep-rl\u002Fmeta-rl-and-generalization","Meta-RL and Generalization","An agent that masters one task often fails on the next; it has overfit to a single environment. This lesson treats fast adaptation as a meta-problem over a distribution of tasks: meta-train so that a few episodes at meta-test time suffice. We cover the two families — optimization-based (MAML learns an initialization) and context-based (RL-squared and PEARL infer a latent task) — the exploration cost of adaptation, and the parallel problem of generalization: why deep RL memorizes environments and what fixes it (domain randomization, procedural generation, augmentation, regularization). It closes on foundation models and sequence-model agents as the generalist endpoint.\n",{"path":17675,"title":17676,"module":17677,"summary":17678},"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fpsychology-of-reinforcement","The Psychology of Reinforcement","Reinforcement Learning in Minds and Brains","Reinforcement learning is both an engineering method and a theory of how animals learn. The prediction\u002Fcontrol split of the algorithms mirrors the psychologist's split between classical and instrumental conditioning. We trace the correspondence: the Rescorla–Wagner model as a prediction-error rule that explains blocking, its real-time TD extension, Thorndike's Law of Effect behind trial-and-error control, and the habitual\u002Fgoal-directed distinction that maps onto model-free versus model-based learning.\n",{"path":17680,"title":17681,"module":17677,"summary":17682},"\u002Freinforcement-learning\u002Fminds-and-brains\u002Finstrumental-conditioning-and-control","The Psychology of Reinforcement: Instrumental Control","Classical conditioning was prediction; instrumental conditioning is control. Thorndike's Law of Effect is trial-and-error control — selection plus association, search plus memory — and Skinner's shaping and schedules are reward engineering. The habitual\u002Fgoal-directed distinction maps onto model-free versus model-based control, dissociated by outcome devaluation and arbitrated by uncertainty. Delayed reinforcement is the credit-assignment problem, and the stimulus traces and secondary reinforcers of animal-learning theory are eligibility traces and value functions.\n",{"path":17684,"title":17685,"module":17677,"summary":17686},"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fdopamine-and-td-error","Dopamine and the TD Error","The TD error was invented as an algorithm; a decade later it turned out to closely describe the firing of the brain's dopamine neurons. We follow Schultz's experiments — dopamine fires at an unpredicted reward, shifts to the earliest predictive cue, and dips below baseline when a predicted reward is withheld — and match each result to the TD error term by term. We then read the basal ganglia as a neural actor–critic with dopamine as its shared training signal, and close on addiction as a hijacking of that signal.\n",{"path":17688,"title":17689,"module":17677,"summary":17690},"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fdopamine-in-the-brain","Dopamine in the Brain: The Neural Actor–Critic","If phasic dopamine is a TD error, where does it go and what does it change? We follow the axons into the basal ganglia, read the corticostriatal synapse as the place where state, action, and error meet, and map the ventral and dorsal striatum onto the critic and the actor of an actor–critic. Addiction becomes a broken cancellation in the same learning signal, and distributional dopamine extends the scalar RPE into a population code.\n",{"path":17692,"title":17693,"module":17677,"summary":17694},"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fanimal-learning-and-cognition","Animal Learning and Cognition","Three classic associative phenomena turn out to be reinforcement-learning mechanisms seen in behavior. Blocking says learning is driven by prediction error, not co-occurrence, and reduces to least-squares regression fitting a collinear feature. Higher-order conditioning and conditioned reinforcement make a value estimate a secondary reinforcer — bootstrapping in an animal. Delayed reinforcement is the credit-assignment problem, and the stimulus traces and goal gradients of Pavlov and Hull are eligibility traces and TD-learned value functions.\n",{"path":17696,"title":17697,"module":17677,"summary":17698},"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fcognitive-maps-and-planning","Cognitive Maps and Model-Based Learning","Tolman's rats learned the layout of a maze with no reward, then used it the moment food appeared — latent learning, a cognitive map, and the behavioral face of model-based reinforcement learning. The map is learned by system identification (stimulus–stimulus associations), which fills in whether or not reward is present, and queried by planning, which re-solves a route from a single changed reward. The successor representation sits between cache and model, and hippocampal predictive maps and scaled-up world models carry the same idea into brain and machine.\n",{"path":17700,"title":17701,"module":17677,"summary":17702},"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fneuroscience-of-reinforcement","The Neuroscience of Reinforcement","The dopamine story is one contact point between reinforcement learning and the brain; this lesson fills in the surrounding neuroscience so the mapping stands on its own. We build a working primer of neurons, synapses, and neuromodulation; separate four signals that casual usage conflates — reward, reinforcement, value, and prediction error; and read the actor and critic as corticostriatal synapses updated by two- and three-factor rules, grounded in spike-timing-dependent and reward-modulated plasticity.\n",{"path":17704,"title":17705,"module":17677,"summary":17706},"\u002Freinforcement-learning\u002Fminds-and-brains\u002Fseveral-learning-systems","The Brain's Several Learning Systems","The actor's three-factor rule has an ancestor in Klopf's hedonistic neuron — a single cell as a reinforcement-seeking agent — and a bacterium's run-and-twiddle shows the Law of Effect with no synapses at all. Teams of such neurons implement policy gradient collectively, the broadcast reward replacing backpropagation. And the brain is not only model-free: outcome devaluation, prefrontal value coding, and hippocampal forward sweeps localize a model-based system. The recurring conclusion is that the brain is several interacting learning systems, not one algorithm.\n",{"path":17708,"title":16864,"module":6,"summary":6},"\u002Freinforcement-learning",{"path":17710,"title":17711,"module":14054,"summary":17712},"\u002Fartificial-intelligence\u002Ffoundations\u002Fwhat-is-ai","What Is Artificial Intelligence?","Eight definitions of AI fall into a two-by-two grid: think versus act, and measure success against human performance versus an ideal standard of rationality. We work through all four schools — the Turing test, cognitive modelling, the laws of thought, and the rational agent — and adopt the last as the frame for the whole course: AI is the study and design of rational agents.\n",{"path":17714,"title":17715,"module":14054,"summary":17716},"\u002Fartificial-intelligence\u002Ffoundations\u002Ffoundations-of-ai","The Foundations of AI","Where the rational-agent idea came from and what surrounds it. AI inherited its core tools from eight older disciplines — philosophy, mathematics, economics, neuroscience, psychology, computer engineering, control theory, and linguistics. Its history runs in cycles of boom and winter, from the 1956 Dartmouth workshop through expert systems to the statistical turn. And the deep-learning era — AlexNet, the Transformer, GPT-3, AlphaGo — is a new way of computing the agent function at scale, not a new definition of AI.\n",{"path":17718,"title":17719,"module":14054,"summary":17720},"\u002Fartificial-intelligence\u002Ffoundations\u002Fintelligent-agents","Intelligent Agents","An agent perceives an environment through sensors and acts on it through actuators; its behavior is an agent function mapping percept sequences to actions. A rational agent chooses, for each percept sequence, the action that maximizes its expected performance measure given its knowledge. We build the first half of the vocabulary the whole course rests on — the agent function, rationality, PEAS task specifications, and the six axes along which task environments vary.\n",{"path":17722,"title":17723,"module":14054,"summary":17724},"\u002Fartificial-intelligence\u002Ffoundations\u002Fagent-architectures","Agent Architectures","How to build a program that computes a good agent function without storing an astronomically large lookup table. Four skeleton architectures in order of increasing power — simple reflex, model-based, goal-based, and utility-based — plus the learning agent that improves any of them, the scale of world representations (atomic, factored, structured) they rest on, and how a modern language-model agent fits the same frame.\n",{"path":17726,"title":17727,"module":17728,"summary":17729},"\u002Fartificial-intelligence\u002Fsearch\u002Funinformed-search","Uninformed Search","Search","A goal-based agent that cannot see which action is best turns the problem into a state space — an initial state, a set of actions, a transition model, a goal test, and a path cost — and searches for a sequence of actions reaching the goal. We build the state-space formulation on the 8-puzzle and route-finding, give the one TREE-SEARCH \u002F GRAPH-SEARCH skeleton every algorithm specializes, and measure strategies by completeness, optimality, and complexity. This lesson develops the first two frontier disciplines — breadth-first and uniform-cost search; the rest follow in the next lesson.\n",{"path":17731,"title":17732,"module":17728,"summary":17733},"\u002Fartificial-intelligence\u002Fsearch\u002Fsearch-strategies-compared","Search Strategies Compared","Breadth-first and uniform-cost search pay for optimality in memory. This lesson develops the strategies that trade memory for depth: depth-first search, which keeps only the current path; depth-limited and iterative-deepening search, which fix DFS's failure on infinite paths; and bidirectional search, which meets in the middle for a square-root saving. It closes by lining up all six uninformed strategies against completeness, optimality, and complexity, and tracing where the algorithms came from and where they went.\n",{"path":17735,"title":17736,"module":17728,"summary":17737},"\u002Fartificial-intelligence\u002Fsearch\u002Finformed-search","Informed Search and A*","An informed search uses a heuristic $h(n)$, an estimate of the cost from a node to the goal, to decide what to expand next. Greedy best-first search follows the heuristic blindly and gives up optimality; A* corrects it by ranking nodes on $f(n) = g(n) + h(n)$, and is optimal when the heuristic is admissible (tree search) or consistent (graph search). This lesson defines the heuristic, builds best-first search, and proves why A* is optimal, with the contour picture that explains its pruning. Where good heuristics come from is the next lesson.\n",{"path":17739,"title":17740,"module":17728,"summary":17741},"\u002Fartificial-intelligence\u002Fsearch\u002Fheuristic-functions","Heuristic Functions and Memory-Bounded Search","A* is only as good as its heuristic, so this lesson answers where good heuristics come from: relaxed problems, whose exact solution cost is an admissible heuristic, and pattern databases, which precompute subproblem costs. It measures heuristic quality with dominance and the effective branching factor, then tackles A*'s memory problem with IDA*, RBFS, and SMA*. It closes with modern heuristic search — weighted A*, learned and disjoint pattern-database heuristics, and bidirectional A*.\n",{"path":17743,"title":17744,"module":17728,"summary":17745},"\u002Fartificial-intelligence\u002Fsearch\u002Flocal-search","Local Search and Optimization","When the path to a goal is irrelevant and only the final state matters, we can discard the search tree entirely and keep just the current state, moving to a better neighbor at each step. This lesson builds the state-space landscape metaphor, works through hill climbing and the three obstacles that defeat it (local maxima, ridges, plateaus), then develops the first escapes: random restarts and simulated annealing with its temperature schedule. The population-based methods and continuous-space calculus follow in the next lesson.\n",{"path":17747,"title":17748,"module":17728,"summary":17749},"\u002Fartificial-intelligence\u002Fsearch\u002Fpopulation-and-continuous-search","Population and Continuous Search","Single-state local search escapes a trap by restarting or tolerating downhill moves. This lesson develops the alternatives that keep several states at once — local beam search, which shares successors across parallel threads, and genetic algorithms, which recombine two parents through crossover and mutation — then crosses into continuous spaces, where calculus replaces the finite neighbor set: gradient ascent, line search, and Newton's method. It closes with the industrial descendants of these methods and the loop they all share.\n",{"path":17751,"title":17752,"module":17728,"summary":17753},"\u002Fartificial-intelligence\u002Fsearch\u002Fadversarial-search","Adversarial Search and Games","When another agent plans against you, search becomes a game. We formalize two-player, zero-sum, perfect-information games as search problems, define the minimax value that optimal play backs up through the game tree, and give the MINIMAX algorithm that computes it. Alpha–beta pruning then cuts the cost of that search roughly in half in the exponent without changing the answer, and a heuristic evaluation function plus a cutoff test turns the exact algorithm into a real-time player that copes with the horizon effect.\n",{"path":17755,"title":17756,"module":17728,"summary":17757},"\u002Fartificial-intelligence\u002Fsearch\u002Fgames-of-chance-and-imperfect-information","Games of Chance and Imperfect Information","Minimax and alpha–beta assume a deterministic game both players can see in full. Drop either assumption and search must change. This lesson adds chance nodes and the expectiminimax value for games with dice, then belief-state reasoning for partially observable games — Kriegspiel and card games — where averaging over clairvoyance both helps and misleads. It closes with the line from Deep Blue's alpha–beta to AlphaGo's learned evaluation and Monte Carlo tree search, and the provable-pruning and self-play research around each end of that story.\n",{"path":17759,"title":17760,"module":17728,"summary":17761},"\u002Fartificial-intelligence\u002Fsearch\u002Fconstraint-satisfaction","Constraint Satisfaction Problems","A constraint satisfaction problem replaces the black-box state with a factored one: variables, domains, and constraints. That structure supports inference before any search runs. This lesson defines the CSP on map coloring, Sudoku, and scheduling, then develops constraint propagation: node and arc consistency, the AC-3 algorithm that makes a whole network arc-consistent, and the way one deleted value cascades across the graph to prune impossible options ahead of search.\n",{"path":17763,"title":17764,"module":17728,"summary":17765},"\u002Fartificial-intelligence\u002Fsearch\u002Fcsp-search-and-structure","CSP Search and Structure","Propagation prunes a CSP but rarely finishes it, so we search. This lesson builds backtracking search over partial assignments and the general-purpose heuristics that make it fast — MRV, degree, least-constraining-value, forward checking, MAC, and intelligent backtracking. It then shows how the shape of the constraint graph controls difficulty: tree-structured problems fall in linear time, cutset conditioning handles the rest, and min-conflicts local search solves a million queens in a constant number of steps.\n",{"path":17767,"title":17768,"module":17728,"summary":17769},"\u002Fartificial-intelligence\u002Fsearch\u002Fsearch-under-uncertainty","Search Under Uncertainty","Classical search assumes the agent knows the state it is in and exactly what each action does. Drop the second assumption and a plan can no longer be a fixed sequence of actions. This lesson develops the first response: AND-OR search over nondeterministic actions, which returns a branching contingency plan rather than a straight line. We build it on the erratic vacuum world, show how OR nodes (the agent's choices) alternate with AND nodes (nature's outcomes), trace the recursion that finds a plan, and handle the case where the only solution is a cyclic \"try, try again.\"\n",{"path":17771,"title":17772,"module":17728,"summary":17773},"\u002Fartificial-intelligence\u002Fsearch\u002Fbelief-state-and-online-search","Belief-State and Online Search","When the agent cannot see the full state, a plan can no longer test where it actually is — it must reason over the set of states it might be in. This lesson develops belief-state search, from sensorless (conformant) planning that coerces an unknown world into a goal, through the predict-observe-update cycle of contingent planning with percepts, to online search in unknown environments, where the agent must act in order to learn. It closes with LRTA*, which refines its own heuristic as it explores, one step from reinforcement learning.\n",{"path":17775,"title":17776,"module":17777,"summary":17778},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fpropositional-logic","Logical Agents and Propositional Logic","Logic and Planning","A knowledge-based agent keeps a store of sentences and acts by asking it what to do. To make \"asking\" mean something we need entailment — the relation $KB \\models \\alpha$ that holds when every model of the knowledge base is a model of the query. Propositional logic gives a syntax and a truth-table semantics for which entailment is decidable. This first part builds the foundations: the agent loop, the Wumpus World, models and entailment, the connectives and truth tables, theorem proving by refutation, and the resolution rule with its CNF conversion — a single complete inference procedure for all of propositional logic.\n",{"path":17780,"title":17781,"module":17777,"summary":17782},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fpropositional-inference","Propositional Inference and Logical Agents","Model checking and resolution decide entailment, but both can blow up. This part turns propositional logic into a practical engine and a working agent. Horn clauses give linear-time forward and backward chaining — the basis of logic programming. DPLL and WalkSAT make satisfiability testing fast in the common case. Then we make the agent situated: time-indexed fluents, the frame problem and its solution by successor-state axioms, a hybrid agent that deduces a safe map and plans a route through it, and SATPlan, which finds a plan by asking a SAT solver for a satisfying model.\n",{"path":17784,"title":17785,"module":17777,"summary":17786},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Ffirst-order-logic","First-Order Logic","Propositional logic can only say that facts hold; it cannot talk about the objects a fact is about, or state a rule once and have it cover every object. First-order logic fixes this by committing to a world of objects, relations, and functions. This first part builds the language from the ground up: the ontology it commits to, the model that gives a sentence a truth value, the syntax of terms and sentences, the two quantifiers with their standard mistakes, and equality.\n",{"path":17788,"title":17789,"module":17777,"summary":17790},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Ffirst-order-logic-in-use","First-Order Logic in Use","With the language of first-order logic in hand, this part is about using it well. Database semantics trades expressive power for the convenience of a single intended model; higher-order logic shows what first-order logic gives up for decidability. Then we put the language to work: the Tell\u002FAsk interface, the kinship domain axiomatized from scratch, and the seven-step knowledge-engineering process applied to a digital circuit.\n",{"path":17792,"title":17793,"module":17777,"summary":17794},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Finference-and-resolution","Inference in First-Order Logic","Propositional inference lifts to first-order logic once we can make terms match. Unification is that machinery: the algorithm that finds the substitution making two expressions identical, and the basis of generalized modus ponens. This first part builds the lifted inference rules and the two chaining algorithms they drive — forward chaining, the data-driven procedure behind production systems and Datalog, and backward chaining, the goal-driven procedure behind Prolog.\n",{"path":17796,"title":17797,"module":17777,"summary":17798},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Ffirst-order-resolution","First-Order Resolution","Chaining is complete only for Horn knowledge bases. General first-order sentences — with disjunctive conclusions and negations — need a single sound and complete rule: resolution. This part converts arbitrary sentences to CNF by skolemizing away the existentials, lifts the resolution rule with unification, and proves entailment by refuting the negated goal. The result is the proof procedure Gödel's completeness theorem guarantees will find any entailment, together with the search strategies that make it usable.\n",{"path":17800,"title":17801,"module":17777,"summary":17802},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fclassical-planning","Classical Planning","Classical planning represents a problem in a factored language, PDDL: states are sets of ground fluents, and actions are lifted schemas with a precondition and an effect. That structure turns planning into search — forward through states or backward through goals — and lets a program read heuristics straight off the schemas by relaxing the problem. This first part develops the representation, the two search directions, and the domain-independent heuristics that come from ignoring preconditions or delete lists.\n",{"path":17804,"title":17805,"module":17777,"summary":17806},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fplanning-graphs-and-graphplan","Planning Heuristics and GraphPlan","Every relaxation heuristic can be inaccurate, and none can tell how far apart subgoals sit. The planning graph is a polynomial-size structure that does better: leveled off the problem, it yields admissible distance estimates and a record of which actions and fluents cannot coexist. This part builds the graph, reads heuristics from it, extracts plans with GraphPlan, and closes with the other classical approaches — SATPlan and partial-order planning — and the representational trade that makes all of it work.\n",{"path":17808,"title":17809,"module":17777,"summary":17810},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fplanning-in-the-real-world","Planning and Acting in the Real World","Classical planning's clean theory rests on four assumptions: time is ignored, actions are atomic, the world is deterministic and fully observable, and the agent is alone. This first part drops the first two. We add durations and resource constraints — turning a plan into a schedule, solved by the critical-path method and, once resources contend, by NP-hard job-shop scheduling — and let a planner reason at multiple levels of abstraction through high-level actions and their angelic reachable sets.\n",{"path":17812,"title":17813,"module":17777,"summary":17814},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fplanning-under-uncertainty","Planning Under Uncertainty","Classical planning assumed the world was deterministic, fully observable, and the agent alone. This part drops the last two assumptions. When the agent cannot see or predict the world, planning moves into belief-state space: sensorless plans that coerce the world into the goal without sensing, contingent plans that branch on what is sensed, and online agents that monitor and replan when execution diverges. Then we add other agents — joint plans, the coordination problem, and the conventions that let a team act without constant negotiation.\n",{"path":17816,"title":17817,"module":17777,"summary":17818},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Fknowledge-representation","Knowledge Representation","First-order logic gives you the language; this lesson is about what to say in it. This first part builds the content: a general upper ontology from the top down, categories as first-class objects with taxonomies and inheritance, physical composition and the count-noun\u002Fmass-noun split, events and time reified through the event calculus, and belief modeled with modal logic — the machinery for representing the world an agent reasons about.\n",{"path":17820,"title":17821,"module":17777,"summary":17822},"\u002Fartificial-intelligence\u002Flogic-and-planning\u002Freasoning-systems-and-defaults","Reasoning Systems and Default Logic","Having represented the world, this part is about reasoning with it at scale. Semantic networks give a graphical notation with fast inheritance; description logics keep subsumption and classification tractable by design. Then we confront the fact that most useful rules hold only by default: circumscription and default logic give a logical account of nonmonotonic reasoning, and truth maintenance systems retract conclusions cleanly when the beliefs beneath them change.\n",{"path":17824,"title":17825,"module":17826,"summary":17827},"\u002Fartificial-intelligence\u002Funcertainty\u002Fprobability-and-bayes","Quantifying Uncertainty","Uncertainty","Logic breaks down in any domain where the rules have exceptions you cannot enumerate — the qualification problem. Probability replaces truth values with degrees of belief that obey Kolmogorov's axioms, and the full joint distribution becomes a knowledge base from which any query is answered by summing entries: marginalization, conditioning, and normalization. Independence factors that joint into smaller pieces — the first step toward a calculus of rational belief that an agent can actually compute with.\n",{"path":17829,"title":17830,"module":17826,"summary":17831},"\u002Fartificial-intelligence\u002Funcertainty\u002Fbayes-rule-and-naive-bayes","Bayes' Rule and Naive Bayes","Bayes' rule inverts a causal model into a diagnostic one, turning \"how a cause produces its symptoms\" into \"which cause explains what I observed.\" Ignoring the prior is the base-rate fallacy behind overconfident test results. Conditional independence then lets several pieces of evidence combine by multiplying likelihood ratios instead of building an exponential joint, giving the naive Bayes model and pointing directly at Bayesian networks.\n",{"path":17833,"title":17834,"module":17826,"summary":17835},"\u002Fartificial-intelligence\u002Funcertainty\u002Fbayesian-networks","Bayesian Networks","A Bayesian network is a directed acyclic graph of random variables in which each node carries a conditional probability table for itself given its parents. That structure factors the full joint distribution into a product of local terms, turning an exponential table into a linear one, and it makes the conditional independences of the domain explicit. We build the canonical burglary–alarm network, read compactness and d-separation off the graph, run exact inference by variable elimination, and, where that is intractable, estimate answers by sampling.\n",{"path":17837,"title":17838,"module":17826,"summary":17839},"\u002Fartificial-intelligence\u002Funcertainty\u002Finference-in-bayesian-networks","Bayesian Networks: Inference and Relational Models","When exact inference is intractable, sampling estimates the posterior instead: prior and rejection sampling, likelihood weighting, and Gibbs\u002FMCMC, whose error shrinks as one over the square root of the sample count. The same graphical idea then lifts from a fixed set of variables to whole populations — relational and open-universe probability models write dependencies once and unroll them over objects — and we close by placing probability against the rule-based, Dempster–Shafer, and fuzzy alternatives it displaced.\n",{"path":17841,"title":17842,"module":17826,"summary":17843},"\u002Fartificial-intelligence\u002Funcertainty\u002Freasoning-over-time","Probabilistic Reasoning over Time","A world that changes needs a state variable at every point in time. The Markov assumption cuts the dependence on history down to the previous slice, leaving a transition model and a sensor model that define a temporal Bayesian network. Four recursive tasks fall out — filtering, prediction, smoothing, and the most likely explanation — each a message passed along the sequence. We ground them in hidden Markov models and their matrix form, sketch the Kalman filter for continuous state, and reach dynamic Bayesian networks with particle filtering as the general approximate method.\n",{"path":17845,"title":17846,"module":17826,"summary":17847},"\u002Fartificial-intelligence\u002Funcertainty\u002Ftracking-and-data-association","Reasoning over Time: Tracking and Data Association","Dynamic Bayesian networks generalize HMMs and Kalman filters to arbitrarily many state variables per slice, and when exact inference blows up, particle filtering approximates the belief state with a population of weighted samples that propagate, reweight, and resample. Tracking several objects at once adds the data-association problem — which observation came from which object — whose combinatorics defeat any exact filter, so particle filters and MCMC keep many hypotheses alive. We close with SLAM and learned state-space models.\n",{"path":17849,"title":17850,"module":17826,"summary":17851},"\u002Fartificial-intelligence\u002Funcertainty\u002Fmaking-decisions","Making Decisions: Utility Theory","A rational agent chooses the action that maximizes expected utility — the probability of each outcome weighted by how much the agent wants it. We derive the utility function from six axioms on preferences, so maximizing expected utility is forced by consistency rather than assumed; look at risk aversion in the utility-of-money curve; package one-shot choices into decision networks; and quantify what an observation is worth with the value of information.\n",{"path":17853,"title":17442,"module":17826,"summary":17854},"\u002Fartificial-intelligence\u002Funcertainty\u002Fmarkov-decision-processes","When an agent must act repeatedly in a stochastic world, a fixed plan is useless — it needs a policy, an action for every state. The Markov decision process makes this precise with a transition model, a reward, and a discount factor; the Bellman equation characterizes the optimal state utilities, and value iteration and policy iteration solve it. Partial observability lifts the problem to belief states, and bandits, Monte-Carlo tree search, and scalable POMDP solvers extend it — this is the model-known half of reinforcement learning.\n",{"path":17856,"title":17857,"module":17826,"summary":17858},"\u002Fartificial-intelligence\u002Funcertainty\u002Fdecision-networks-and-game-theory","Decision Analysis: Multi-Attribute Utility and Decision Networks","Decision analysis takes the single-agent utility framework and makes it practical: utility over several attributes, dominance and additive value functions, influence diagrams that fold Bayesian networks together with decision and utility nodes, and the value of information that tells an agent which questions are worth asking. Structure in an agent's preferences — dominance, preferential and utility independence — collapses an exponential utility table into a few one-dimensional functions, the same move that made Bayesian networks compact.\n",{"path":17860,"title":17861,"module":17826,"summary":17862},"\u002Fartificial-intelligence\u002Funcertainty\u002Fgame-theory-and-mechanism-design","Game Theory and Mechanism Design","When outcomes depend on other rational agents, single-agent utility maximization no longer suffices. Game theory studies decisions among agents — normal-form games, dominant strategies, Nash and maximin equilibria, and repeated games — and mechanism design runs the logic backwards, engineering rules (auctions, VCG) so that self-interested play produces a good collective outcome. Algorithmic game theory then asks whether equilibria can be computed, what selfishness costs society, and how the mechanisms deployed at internet scale actually behave.\n",{"path":17864,"title":17865,"module":17866,"summary":17867},"\u002Fartificial-intelligence\u002Flearning\u002Flearning-from-examples","Learning from Examples","Learning","An agent that improves with experience does not need its designer to anticipate every situation. Inductive learning takes that ambition and narrows it to one tractable problem: from labelled input-output pairs, recover a function that predicts the output for inputs never seen. This first part builds the foundation around a single organizing question — generalization — through decision trees and information gain, and the training\u002Fvalidation\u002Ftest discipline for evaluating and choosing hypotheses. A second part takes up the theory of learning and the main model families.\n",{"path":17869,"title":17870,"module":17866,"summary":17871},"\u002Fartificial-intelligence\u002Flearning\u002Ftheory-and-model-families","The Theory of Learning and Model Families","Cross-validation measures generalization but does not explain it. This part supplies the theory — PAC learning, sample complexity, and the VC dimension — that says when a hypothesis consistent with enough data is probably approximately correct, and why an unrestricted hypothesis space can never generalize. It then surveys the model families a practitioner reaches for: linear regression and gradient descent, the perceptron and logistic regression, support vector machines and the kernel trick, and ensembles by bagging and boosting — closing with what deep learning changed about the classical picture.\n",{"path":17873,"title":17874,"module":17866,"summary":17875},"\u002Fartificial-intelligence\u002Flearning\u002Fprobabilistic-learning","Learning Probabilistic Models","A [Bayesian network](\u002Fartificial-intelligence\u002Funcertainty\u002Fbayesian-networks) is useless until its numbers are filled in, and those numbers come from data. This first part casts learning itself as probabilistic inference: hypotheses carry a prior, data update it to a posterior, and predictions average over what remains. From that frame fall the standard estimators — maximum likelihood by counting, MAP with a conjugate prior, full Bayesian updating — for the case where every variable is observed. A second part takes up the harder case of hidden variables and the EM algorithm.\n",{"path":17877,"title":17878,"module":17866,"summary":17879},"\u002Fartificial-intelligence\u002Flearning\u002Fexpectation-maximization","Learning with Hidden Variables: The EM Algorithm","Complete data can be learned by counting; real data usually hide some variables — the disease behind the symptoms, the cluster behind the points. This part develops the expectation-maximization algorithm, which learns those models by alternating an expected completion of the missing data with a re-estimation of the parameters. It works the idea through mixtures of Gaussians, Bayesian networks, and hidden Markov models, proves the monotone-likelihood guarantee from the evidence lower bound, and traces the line from EM to variational inference and the variational autoencoder.\n",{"path":17881,"title":16864,"module":17866,"summary":17882},"\u002Fartificial-intelligence\u002Flearning\u002Freinforcement-learning","Reinforcement learning is an MDP with the model unknown: the agent knows neither how its actions move the world nor which states are rewarded, and must recover good behaviour from experienced transitions and rewards alone. This first part builds the classical tabular theory — passive learning (fix a policy, learn its value, by direct estimation, adaptive dynamic programming, and temporal differences) and active learning (choose actions, trade exploration against exploitation, and learn control with Q-learning and SARSA). A second part lifts it off the lookup table with function approximation and policy search.\n",{"path":17884,"title":17885,"module":17866,"summary":17886},"\u002Fartificial-intelligence\u002Flearning\u002Fgeneralization-and-policy-search","Reinforcement Learning: Generalization and Policy Search","Tabular reinforcement learning stores one number per state, which is hopeless for backgammon or chess. This part lifts RL off the lookup table with function approximation, so that updating one state generalizes to related ones, then turns to policy search — representing and optimizing the policy directly, up to the REINFORCE policy gradient and correlated sampling. It closes with the bridge to deep reinforcement learning (deep Q-networks, actor-critic, PPO), the classic applications, and the hand-off to the dedicated RL subject.\n",{"path":17888,"title":17889,"module":17866,"summary":17890},"\u002Fartificial-intelligence\u002Flearning\u002Fknowledge-in-learning","Knowledge in Learning","Pure induction learns a function from labelled examples while knowing almost nothing to begin with. This first part brings prior knowledge into the loop by recasting learning as logical inference — hypotheses, examples, and classifications as sentences. It develops current-best-hypothesis search, the version space and its general\u002Fspecific boundary maintained by candidate elimination, and states the three entailment constraints that fix how background knowledge enters. A second part builds the three knowledge-based methods those constraints define.\n",{"path":17892,"title":17893,"module":17866,"summary":17894},"\u002Fartificial-intelligence\u002Flearning\u002Fknowledge-based-learning-methods","Knowledge-Based Learning: EBL, Relevance, and ILP","Once learning is cast as logical inference, three methods follow from the three ways prior knowledge can enter. Explanation-based learning generalizes a single example by explaining it with the domain theory, gaining speed but nothing new. Relevance-based learning uses determinations to shrink the hypothesis space and converge from fewer examples. Inductive logic programming learns genuinely new first-order rules — top-down with FOIL, bottom-up by inverting resolution, even inventing new predicates — and connects to modern statistical relational and neuro-symbolic learning.\n",{"path":17896,"title":17897,"module":17898,"summary":17899},"\u002Fartificial-intelligence\u002Ffrontiers\u002Fvision-and-perception","Vision and Perception","Frontiers","Perception connects an agent to the physical world. We follow one modality — vision — from the physics of image formation (the pinhole camera, perspective projection, lenses, shading, color) through the early operations that turn a pixel array into edges, texture, and motion, and into recognition by appearance. The recurring problem is inversion: a camera collapses a 3-D world onto a 2-D grid, and an agent that wants to act must build the scene back up. Rebuilding the scene is the subject of the companion lesson.\n",{"path":17901,"title":17902,"module":17898,"summary":17903},"\u002Fartificial-intelligence\u002Ffrontiers\u002Freconstructing-the-3d-world","Vision: Reconstructing the 3D World","A camera collapses a three-dimensional world onto a flat grid; this lesson inverts that collapse. We build the camera projection matrix (intrinsics and extrinsics), triangulate a point from two views, then work through the toolbox of depth cues — motion parallax, binocular stereopsis, multiple views, texture, shading, and contour — that turn an ambiguous image back into a scene. We add structural recognition (pictorial-structure \"cardboard people\"), the task-driven use of vision in cars and robots, and the shift from hand-built pipelines to learned deep-vision networks.\n",{"path":17905,"title":17906,"module":17898,"summary":17907},"\u002Fartificial-intelligence\u002Ffrontiers\u002Frobotics","Robotics","A robot is an agent with a body: sensors that read the physical world and effectors that push back on it. This lesson grounds the abstract AI machinery in that body. We build up the hardware (range finders, proprioception, degrees of freedom), then cast perception as probabilistic filtering — the kinematic motion and sensor models, Monte Carlo localization, the extended Kalman filter, and simultaneous localization and mapping (SLAM). The companion lesson takes the estimated pose forward into planning and control.\n",{"path":17909,"title":17910,"module":17898,"summary":17911},"\u002Fartificial-intelligence\u002Ffrontiers\u002Frobot-planning-and-control","Robotics: Planning and Control","A robot that knows where it is still has to decide how to move, and then make a slipping, sensing-imperfect body actually go there. This lesson takes the pose estimate forward: planning motion in configuration space with cell decomposition and sampling-based roadmaps (PRMs and RRTs), planning under uncertainty with most-likely-state and online replanning, closing the loop with P\u002FPD\u002FPID control and potential fields, and finally the software architectures — subsumption, three-layer, and pipeline — that assemble it all, plus the learning-based turn in modern robotics.\n",{"path":17913,"title":17914,"module":17898,"summary":17915},"\u002Fartificial-intelligence\u002Ffrontiers\u002Fnatural-language-in-ai","Natural Language for AI Agents","Language is how agents acquire the knowledge already written down and how they communicate with the humans they serve. This lesson gives the classical AI account of language as a source of information: n-gram language models and the information-seeking tasks built on them — text classification, information retrieval (BM25, the inverted index, PageRank), and information extraction with finite-state templates and hidden Markov models. Throughout, we point to the dedicated NLP subject for the modern deep-learning treatment; the companion lesson takes up grammar, translation, and speech.\n",{"path":17917,"title":17918,"module":17898,"summary":17919},"\u002Fartificial-intelligence\u002Ffrontiers\u002Fnlp-grammar-translation-and-speech","Language for AI Agents: Grammar, Translation, and Speech","N-gram models see only a local window; they cannot say why \"black dog\" is well-formed English and \"dog black\" is not, because that is a fact about structure. This lesson takes up structure: phrase-structure and probabilistic context-free grammars, syntactic analysis by chart parsing and CYK, augmented grammars and compositional semantics, then the two major statistical successes — machine translation and speech recognition — cast as noisy-channel problems. It closes with the bridge from n-grams to transformers and where the classical account sits relative to modern NLP.\n",{"path":17921,"title":17922,"module":17898,"summary":17923},"\u002Fartificial-intelligence\u002Ffrontiers\u002Fphilosophy-and-future","Philosophy, Ethics, and the Future of AI","Two questions have shadowed the field since its founding: can machines act intelligently (weak AI), and can they really think (strong AI)? We work through Turing's objections and their rebuttals — the arguments from disability, mathematics, and informality — then the strong-AI debate: the mind-body problem, functionalism and the brain prosthesis, Searle's Chinese Room and the systems reply, and consciousness and qualia. The companion lesson turns from what AI can do to what it should, and closes the course.\n",{"path":17925,"title":17926,"module":17898,"summary":17927},"\u002Fartificial-intelligence\u002Ffrontiers\u002Fai-ethics-and-future","The Ethics and Future of AI","Having asked whether machines can act intelligently and really think, we turn to whether we should build them at all. This lesson works through the six ethical risks — lost jobs, autonomous weapons, surveillance and privacy, biased decisions, the safety of superintelligence, and the erosion of accountability — then the value-alignment problem in the LLM era, and where the classical agent components could go next. It closes the course by tying search, logic, probability, and learning into a single picture of intelligence as rational agency.\n",{"path":17929,"title":17930,"module":6,"summary":6},"\u002Fartificial-intelligence","Artificial Intelligence",{"path":17932,"title":17933,"module":17934,"summary":17935},"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-constituents-nuclide-chart","Nuclear Composition and Ground-State Properties","Nuclear Properties","The nucleus is a bound assembly of Z protons and N neutrons packed to a radius R = R0 A^(1\u002F3) at a nearly constant density of about 10^17 kg\u002Fm^3. We fix the vocabulary of nuclides, derive nuclear size from mirror-nuclide and electron-scattering data, read the binding-energy-per-nucleon curve, and model it with the liquid-drop semiempirical mass formula.\n",{"path":17937,"title":17938,"module":17934,"summary":17939},"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-size-charge-distributions","Nuclear Size, Shape, and Charge Distributions","Elastic electron scattering resolves the nucleus by its de Broglie wavelength. The measured cross section is the Mott point-charge cross section modulated by a form factor, and that form factor is the Fourier transform of the charge density. Diffraction minima fix the radius, the small-angle slope fixes the mean-square radius, and the fitted Woods-Saxon profile gives a central density and a skin thickness. Mirror-nucleus Coulomb energies, muonic-atom X-rays, and optical isotope shifts give independent radii that all track R = R0 A^(1\u002F3).\n",{"path":17941,"title":17942,"module":17934,"summary":17943},"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-masses-binding-energy","Nuclear Masses, Mass Excess, and Separation Energies","The atomic mass unit fixes the scale, and the mass excess collects the small binding-driven deviation from the integer mass number. Penning-trap cyclotron frequencies now measure masses to parts in a billion, and every decay and reaction Q-value is a difference of these masses. One- and two-nucleon separation energies read the binding difference between neighbouring nuclides directly, showing the even-odd pairing stagger and the sharp drops at magic numbers, and their vanishing marks the neutron and proton drip lines that bound the chart of the nuclides.\n",{"path":17945,"title":17946,"module":17934,"summary":17947},"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fsemi-empirical-mass-formula","The Semi-Empirical Mass Formula and the Valley of Stability","Five physical terms reproduce nuclear binding across the chart: a volume term from saturation, a surface term from the deficit of edge neighbours, a Coulomb term from the electrostatic self-energy of a charged sphere, an asymmetry term from the Pauli cost of unequal proton and neutron filling, and a pairing term. The formula is quadratic in Z at fixed A, so isobars lie on a mass parabola whose minimum sets the most stable charge and whose slope dictates the direction of beta decay. The same competition between surface and Coulomb energy defines the fissility parameter and the onset of fission.\n",{"path":17949,"title":17950,"module":17934,"summary":17951},"\u002Fnuclear-physics\u002Fnuclear-properties\u002Fnuclear-moments-multipoles","Nuclear Spin, Magnetic Dipole, and Electric Quadrupole Moments","The ground state of a nucleus carries a definite spin and parity, a magnetic dipole moment of order the nuclear magneton, and, when its spin exceeds one-half, an electric quadrupole moment that measures its shape. The single-particle Schmidt lines predict the magnetic moment of an odd-A nucleus from the last unpaired nucleon, and the measured moments fall between them. The quadrupole moment distinguishes prolate from oblate deformation, and hyperfine structure is the experimental handle that fixes the spin and the moments from an atomic spectrum.\n",{"path":17953,"title":17954,"module":17955,"summary":17956},"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fnuclear-force-shell-overview","The Nuclear Force and the Shell Model","The Nuclear Force","The strong force between nucleons is short-range, charge-independent, saturated, and repulsive at its core, about a hundred times stronger than Coulomb. Yukawa explained it as an exchange of massive mesons, tying the force's range to the meson mass through the uncertainty principle. Layered on top, an independent-particle shell model with strong spin-orbit coupling reproduces the magic numbers 2, 8, 20, 28, 50, 82, 126.\n",{"path":17958,"title":17959,"module":17955,"summary":17960},"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fthe-deuteron","The Deuteron and the Tensor Force","The deuteron is the only bound two-nucleon state: one shallow level at 2.22 MeV, no excited states. A square-well fit fixes a depth near 35 MeV over a 2 fm range, yet the wavefunction leaks so far past the edge that most of the probability lies outside the force. Its spin-1 ground state, magnetic moment close to the sum of the free-nucleon moments, and small but nonzero electric quadrupole moment together force a D-state admixture and a non-central tensor force.\n",{"path":17962,"title":17963,"module":17955,"summary":17964},"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fnucleon-nucleon-scattering","Nucleon-Nucleon Scattering and the Interaction's Structure","Scattering probes the nuclear force above threshold. Partial-wave analysis reduces low-energy data to a single s-wave phase shift, and the effective-range expansion packages that into a scattering length and an effective range. The triplet channel binds (the deuteron) while the singlet is only virtual, which together explain the anomalously large free neutron-proton cross section. Comparing pp, nn, and np results establishes charge symmetry and charge independence, and polarization experiments expose the spin-orbit and tensor pieces.\n",{"path":17966,"title":17967,"module":17955,"summary":17968},"\u002Fnuclear-physics\u002Fnuclear-force-deuteron\u002Fmeson-theory-isospin","Meson Exchange, the Yukawa Potential, and Isospin","Yukawa's massive-field propagator turns the range of the nuclear force into a meson mass: the exchanged quantum's Compton wavelength is the range. One-pion exchange fixes the long-range tail, complete with the tensor structure the deuteron demanded, while heavier mesons build the intermediate attraction and the repulsive core. Charge independence becomes an isospin symmetry, the force is diagonalized by the total isospin through a tau-dot-tau interaction, and the whole picture sits inside QCD as a residual color force between color-neutral nucleons.\n",{"path":17970,"title":17971,"module":17972,"summary":17973},"\u002Fnuclear-physics\u002Fnuclear-models\u002Ffermi-gas-model","The Fermi Gas Model","Nuclear Models","Treating the nucleus as two degenerate Fermi gases of protons and neutrons confined in a common well fixes the Fermi momentum near 250 MeV\u002Fc and the Fermi energy near 33 MeV from the nuclear density alone. The average kinetic energy per nucleon is about 20 MeV, the well depth is the Fermi energy plus the separation energy, and unequal proton and neutron Fermi levels reproduce the asymmetry term of the mass formula.\n",{"path":17975,"title":17976,"module":17972,"summary":17977},"\u002Fnuclear-physics\u002Fnuclear-models\u002Fliquid-drop-collective-coordinates","The Liquid-Drop Model and Collective Deformation","Deforming a charged liquid drop into a spheroid raises its surface energy and lowers its Coulomb energy; the two effects compete through the deformation parameter to set a stability minimum and a fission barrier. The ratio of Coulomb to twice the surface energy is the fissility Z-squared over A, which crosses one near 49 and marks the point where the sphere is unstable. The same surface tension that restores small deformations quantizes into collective vibrations, carrying the static mass formula into dynamic collective motion.\n",{"path":17979,"title":17980,"module":17972,"summary":17981},"\u002Fnuclear-physics\u002Fnuclear-models\u002Fshell-model-single-particle","The Shell Model: Single-Particle States and Spin-Orbit Coupling","A harmonic-oscillator well reproduces the first three magic numbers but fails above twenty; adding a strong inverted spin-orbit term that drives the stretched j equals l plus one-half level down closes the gaps at 28, 50, 82, and 126. The filled shells couple to zero, so the last unpaired nucleon fixes the ground-state spin and parity, and its single-particle magnetic moment falls on the Schmidt lines. Configuration mixing sets the limits of the extreme single-particle model.\n",{"path":17983,"title":17984,"module":17972,"summary":17985},"\u002Fnuclear-physics\u002Fnuclear-models\u002Fcollective-model-rotations-vibrations","The Collective Model: Rotations, Vibrations, and Deformed Nuclei","Deformed nuclei rotate with energies proportional to I times I plus one, giving the ground-state band its characteristic level ratios, while near-spherical nuclei vibrate in quantized surface phonons that build one- and two-phonon multiplets. The Nilsson model tracks single-particle levels as the well deforms, moments of inertia fall between the rigid and irrotational limits, backbending marks the sudden alignment of a broken pair, and giant resonances are the bulk dipole and quadrupole modes of the whole nucleus.\n",{"path":17987,"title":17988,"module":17989,"summary":17990},"\u002Fnuclear-physics\u002Fradioactive-decay\u002Fdecay-law-modes","Radioactivity and Decay Modes","Radioactive Decay","Unstable nuclei decay at a rate proportional to how many remain, giving the exponential law N(t) = N0 e^(-lambda t) with half-life t = 0.693\u002Flambda. We work through the three common modes: alpha decay as Coulomb-barrier tunneling with the Geiger-Nuttall rule, beta decay whose continuous spectrum demands the neutrino, and gamma de-excitation, and follow a decay chain across the chart of nuclides.\n",{"path":17992,"title":17993,"module":17989,"summary":17994},"\u002Fnuclear-physics\u002Fradioactive-decay\u002Fdecay-kinetics-equilibrium","Serial Decay, the Bateman Equations, and Radioactive Equilibrium","A radioactive parent that decays into a radioactive daughter obeys a coupled pair of rate equations whose solution is the Bateman formula. Depending on the half-life ordering the chain settles into secular equilibrium (equal activities), transient equilibrium (a fixed activity ratio), or no equilibrium. Constant production under irradiation drives the activity toward a saturation value equal to the production rate, competing decay modes split the total decay constant into partial constants, and the natural decay series in secular equilibrium underpin radiometric dating.\n",{"path":17996,"title":17997,"module":17998,"summary":17999},"\u002Fnuclear-physics\u002Falpha-decay\u002Falpha-decay-gamow-theory","Alpha Decay and the Gamow Theory of Tunneling","Alpha Decay","The alpha Q-value turns positive above mass number 150 because the emitted helium-4 is exceptionally tightly bound. Emission proceeds by quantum tunneling through the Coulomb barrier: a WKB integral from the nuclear surface to the outer turning point gives the Gamow factor, and multiplying its penetrability by the assault frequency yields half-lives spanning more than twenty orders of magnitude. The leading term reproduces the Geiger-Nuttall relation, log t½ proportional to the daughter charge over the square root of Q.\n",{"path":18001,"title":18002,"module":17998,"summary":18003},"\u002Fnuclear-physics\u002Falpha-decay\u002Falpha-fine-structure-hindrance","Fine Structure, Angular Momentum, and Hindrance Factors","A single parent emits several alpha groups of slightly different energy, each feeding a distinct level of the daughter, so the alpha spectrum maps the daughter's low-lying states. Emission with orbital angular momentum L raises the barrier by a centrifugal term and is allowed only when angular-momentum and parity selection rules permit. Comparing the measured partial half-life to the Gamow estimate defines a hindrance factor near unity for even-even ground-state transitions and large for odd-A decays that must rearrange the unpaired nucleon.\n",{"path":18005,"title":18006,"module":18007,"summary":18008},"\u002Fnuclear-physics\u002Fbeta-decay\u002Fbeta-decay-energetics-neutrino","Beta Decay Energetics and the Neutrino","Beta Decay and the Weak Interaction","Beta decay converts a neutron into a proton or the reverse, adjusting Z at fixed A along an isobaric mass parabola. We write the three processes (beta-minus, beta-plus, electron capture), reduce every Q-value to a difference of neutral atomic masses, and read the continuous electron spectrum as the fingerprint of a third, nearly massless particle. Pauli's neutrino, its detection by Reines and Cowan, and the endpoint bound on its mass close the lesson.\n",{"path":18010,"title":18011,"module":18007,"summary":18012},"\u002Fnuclear-physics\u002Fbeta-decay\u002Ffermi-theory-beta-decay","Fermi's Theory: Kurie Plots and ft Values","Fermi treated beta decay as a point-contact weak transition and read its rate from the golden rule. The electron spectrum then follows from phase space and the Coulomb Fermi function; the Kurie plot straightens it to a line whose intercept is the endpoint. Integrating the spectrum gives the comparative half-life ft, whose logarithm sorts transitions into superallowed, allowed, and forbidden classes governed by the Fermi and Gamow-Teller selection rules.\n",{"path":18014,"title":18015,"module":18007,"summary":18016},"\u002Fnuclear-physics\u002Fbeta-decay\u002Fweak-interaction-parity-violation","The Weak Interaction and Parity Violation","Beta decay violates mirror symmetry. The Wu experiment on polarized cobalt-60 showed electrons emitted preferentially against the nuclear spin, a pseudoscalar correlation forbidden if parity were conserved. The result fixes the weak charged current as left-handed V minus A, forces neutrinos to be left-handed and antineutrinos right-handed (measured by Goldhaber), and places beta decay within the electroweak theory as W-boson exchange turning a down quark into an up quark.\n",{"path":18018,"title":18019,"module":18007,"summary":18020},"\u002Fnuclear-physics\u002Fbeta-decay\u002Fdouble-beta-decay-neutrino-mass","Double Beta Decay and Neutrino Mass","For even-A isobars the pairing term splits the mass parabola into two curves, and a handful of even-even nuclides sit below their odd-odd neighbor yet above the next even-even one: single beta decay is forbidden but second-order double beta decay is allowed. The two-neutrino mode is a standard-model process with the longest measured lifetimes in nature; the neutrinoless mode would require the neutrino to be its own antiparticle and its rate measures the effective Majorana mass, the sharpest probe of the absolute neutrino mass scale.\n",{"path":18022,"title":18023,"module":18024,"summary":18025},"\u002Fnuclear-physics\u002Fgamma-decay\u002Fgamma-multipole-radiation","Multipole Radiation and Selection Rules","Gamma Decay","Gamma decay carries a nucleus from an excited state to a lower one by emitting a photon of definite angular momentum and parity. We correct the photon energy for nuclear recoil, expand the radiation field into electric and magnetic multipoles, and read off how the transition rate collapses with each increase in multipole order. The Weisskopf single-particle estimates set the scale, and angular-momentum and parity conservation fix which multipole dominates.\n",{"path":18027,"title":18028,"module":18024,"summary":18029},"\u002Fnuclear-physics\u002Fgamma-decay\u002Finternal-conversion-isomers","Internal Conversion and Isomers","A nucleus can shed excitation energy without emitting a photon by handing it directly to an atomic electron. We define the internal-conversion coefficient, trace its growth with atomic number, multipole order, and decreasing energy, and treat the electron-only E0 transitions and internal pair formation. When the lowest allowed multipole is high and the energy low, the gamma rate falls so far that the excited state survives as a metastable isomer.\n",{"path":18031,"title":18032,"module":18024,"summary":18033},"\u002Fnuclear-physics\u002Fgamma-decay\u002Fangular-correlations-mossbauer","Angular Correlations and the Mössbauer Effect","Two gammas emitted in cascade are not independent in direction: detecting the first selects magnetic substates of the intermediate level and makes the second anisotropic, so the correlation function fixes the intermediate spin. The same nuclear resonance that recoil normally destroys is recovered when the emitter is locked in a lattice, giving the Mössbauer effect and its part-in-a-trillion resolution of isomer shifts and hyperfine fields.\n",{"path":18035,"title":18036,"module":18037,"summary":18038},"\u002Fnuclear-physics\u002Fnuclear-reactions\u002Freaction-kinematics-cross-sections","Nuclear Reactions, Fission, and Fusion","Nuclear Reactions","A nuclear reaction X(x, y)Y is governed by its Q value and its cross section, the effective target area for a given process. Splitting the curve of binding energy near iron in either direction releases energy: fission of heavy nuclei by neutron capture and a chain reaction, and fusion of light nuclei that powers the Sun and needs Lawson's density-confinement criterion to be practical.\n",{"path":18040,"title":18041,"module":18037,"summary":18042},"\u002Fnuclear-physics\u002Fnuclear-reactions\u002Fcompound-nucleus-resonances","The Compound Nucleus and Resonance Reactions","Low-energy reactions proceed through a long-lived intermediate state whose decay forgets how it formed. Bohr's independence hypothesis factorizes the cross section into a formation step and a branching ratio, an isolated level gives the single-level Breit-Wigner line shape with total width Γ tied to the lifetime by Γτ = ħ, and at high excitation overlapping levels merge into a statistical continuum described by evaporation spectra and the Hauser-Feshbach average.\n",{"path":18044,"title":18045,"module":18037,"summary":18046},"\u002Fnuclear-physics\u002Fnuclear-reactions\u002Fdirect-reactions-optical-model","Direct Reactions and the Optical Model","A complex optical potential replaces the many-body target by a single particle moving in an average field whose imaginary part removes flux into non-elastic channels, reproducing the diffraction pattern of elastic scattering. Direct reactions bypass the compound nucleus, transferring a nucleon in one step: stripping and pickup deposit or remove a single nucleon, the angle of the first peak in the distorted-wave angular distribution fixes the transferred orbital angular momentum, and its magnitude gives the spectroscopic factor.\n",{"path":18048,"title":18049,"module":18050,"summary":18051},"\u002Fnuclear-physics\u002Ffission\u002Ffission-barrier-dynamics","The Fission Barrier and Fragment Energetics","Nuclear Fission","Fission is the large-amplitude collective deformation of a heavy nucleus into two fragments. The liquid-drop model sets a barrier from the competition between rising surface energy and falling Coulomb energy under quadrupole deformation, with the fissility parameter Z²\u002FA measuring how close a nucleus is to instability. Bohr-Wheeler theory separates spontaneous from neutron-induced fission, the fragment mass yield is double-humped and asymmetric, about 200 MeV is released per event, and shell corrections add a second minimum that produces fission isomers.\n",{"path":18053,"title":18054,"module":18050,"summary":18055},"\u002Fnuclear-physics\u002Ffission\u002Fchain-reactions-reactor-physics","Chain Reactions and Reactor Physics","A self-sustaining chain reaction is a fixed point of neutron bookkeeping: the multiplication factor k counts the neutrons in one generation per neutron in the last, and criticality is k = 1. The four-factor formula tracks a neutron through fast fission, resonance escape, thermal utilization, and reproduction; moderation slows fission neutrons to the thermal energies where the fission cross section is largest; and the small delayed-neutron fraction sets the timescale that makes a reactor controllable. Breeding converts fertile U-238 and Th-232 into new fissile fuel.\n",{"path":18057,"title":18058,"module":18059,"summary":18060},"\u002Fnuclear-physics\u002Ffusion-nucleosynthesis\u002Ffusion-reactions-confinement","Fusion Reactions and Confinement","Fusion and Nucleosynthesis","Light nuclei release energy when they fuse because binding per nucleon rises steeply toward the iron peak, but the Coulomb barrier suppresses the rate at reactor temperatures. The thermonuclear rate is a convolution of the Maxwell distribution with the tunneling probability, sharply peaked at the Gamow energy. The deuterium-tritium reaction has the lowest barrier and largest cross section; sustained energy gain requires the Lawson triple product of density, temperature, and confinement time, reached by magnetic or inertial confinement.\n",{"path":18062,"title":18063,"module":18059,"summary":18064},"\u002Fnuclear-physics\u002Ffusion-nucleosynthesis\u002Fstellar-nucleosynthesis","Stellar Nucleosynthesis","Main-sequence stars burn hydrogen to helium through the proton-proton chain and the CNO cycle, both releasing 26.7 MeV per helium nucleus. Helium burning bridges the mass-5 and mass-8 gaps by the triple-alpha process through the Beryllium-8 and Hoyle resonances, and successive carbon-to-silicon burning stages climb to the iron peak, where fusion stops. The elements beyond iron are built by slow and rapid neutron capture, and the solar neutrino flux confirms the reactions directly.\n",{"path":18066,"title":18067,"module":18059,"summary":18068},"\u002Fnuclear-physics\u002Ffusion-nucleosynthesis\u002Fbig-bang-nucleosynthesis","Big-Bang Nucleosynthesis","In the first three minutes the expanding universe forged the light elements. The weak interaction froze the neutron-to-proton ratio near one in six when the reaction rate fell below the expansion rate, and free-neutron decay lowered it to about one in seven before the deuterium bottleneck broke. Almost every surviving neutron ended in helium-4, fixing the primordial helium mass fraction near 0.25, with trace deuterium, helium-3, and lithium-7. The deuterium abundance measures the cosmic baryon density.\n",{"path":18070,"title":18071,"module":18072,"summary":18073},"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fcharged-particle-stopping-power","Stopping Power and the Range of Charged Particles","Radiation and Applications","A heavy charged particle loses energy in a dense sequence of small Coulomb collisions with atomic electrons, at a rate the Bethe-Bloch formula fixes from the particle's charge and speed and the medium's electron density and mean excitation energy. The rate scales as the inverse square of the speed, so most energy is deposited at the end of the track in the Bragg peak, and integrating the reciprocal rate gives a sharp range. Electrons differ: they also radiate, and above a critical energy bremsstrahlung dominates. Fast particles above the phase velocity of light in the medium emit Cherenkov radiation.\n",{"path":18075,"title":18076,"module":18072,"summary":18077},"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fphoton-neutron-interactions","Interactions of Photons and Neutrons","Photons are removed from a beam in single events, so their intensity falls exponentially with a linear attenuation coefficient built from three processes: the photoelectric effect at low energy, Compton scattering at intermediate energy, and pair production above twice the electron rest energy, each with its own atomic-number and energy dependence. Neutrons carry no charge and interact only with nuclei, moderating by elastic scattering and being captured with a cross section that rises as one over speed away from resonances.\n",{"path":18079,"title":18080,"module":18072,"summary":18081},"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fradiation-detectors","Radiation Detectors and Nuclear Spectroscopy","Every detector converts the energy a radiation deposits into a measurable electrical signal. Gas counters read the ionization directly, in three operating regions set by the applied voltage; scintillators convert the energy to light read out by a photomultiplier; semiconductor detectors collect electron-hole pairs and give the best energy resolution because so many carriers are made per event. The resolution is governed by the number of independent charge carriers, and the pulse-height spectrum of a gamma line shows a full-energy photopeak, a Compton continuum with its edge, and escape peaks.\n",{"path":18083,"title":18084,"module":18072,"summary":18085},"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fdosimetry-radiation-biology","Dosimetry, Radiation Biology, and Protection","Absorbed dose is the energy deposited per unit mass, measured in gray. Equal absorbed doses do unequal biological damage because densely ionizing radiation deposits its energy along short tracks: weighting the dose by a radiation factor gives the equivalent dose, and weighting by tissue sensitivity gives the effective dose, both in sieverts. Deterministic effects have a threshold and a severity that grows with dose; stochastic effects are assumed to follow a linear-no-threshold probability. Natural background dominates the dose to the population, and protection rests on time, distance, and shielding.\n",{"path":18087,"title":18088,"module":18072,"summary":18089},"\u002Fnuclear-physics\u002Fradiation-matter-applications\u002Fnuclear-applications-dating-medicine","Applications — Dating, Analysis, and Nuclear Medicine","Charged particles lose energy continuously and stop at a well-defined range with a Bragg peak, while gamma rays are attenuated exponentially. These interactions define radiation detectors and dosimetry (gray and sievert) and drive the applications: neutron activation analysis, magnetic resonance imaging, PET, and radiometric dating with carbon-14 and long-lived rock clocks.\n",{"path":18091,"title":18092,"module":6,"summary":6},"\u002Fnuclear-physics","Nuclear Physics",{"path":18094,"title":18095,"module":14054,"summary":18096},"\u002Fnatural-language-processing\u002Ffoundations\u002Fwhat-is-nlp","What Is Natural Language Processing?","Natural language processing is the computational treatment of human language: reading it, representing it, and generating it. We set up why the problem is hard — ambiguity at every level, from sound to intent — trace the field from ELIZA's pattern-matching through statistical methods to today's neural models, lay out the linguistic levels and task families the course covers, and fix the vocabulary of tokens, types, and corpora the rest of the notes rely on.\n",{"path":18098,"title":18099,"module":14054,"summary":18100},"\u002Fnatural-language-processing\u002Ffoundations\u002Fregex-and-text-normalization","Regular Expressions and Text Normalization","Before any model touches text, the text has to be found and cleaned. Regular expressions give an algebra for describing string patterns; tokenization, case folding, and stemming turn raw characters into the units a model counts; and byte-pair encoding builds a subword vocabulary that spells out any word. Measuring how far apart two strings are — minimum edit distance — is the next lesson.\n",{"path":18102,"title":18103,"module":14054,"summary":18104},"\u002Fnatural-language-processing\u002Ffoundations\u002Fminimum-edit-distance","Minimum Edit Distance","Much of language processing needs to measure how similar two strings are — a speller ranking corrections, a diff tool, a coreference resolver. Minimum edit distance counts the insertions, deletions, and substitutions that turn one string into another, computed by a dynamic-programming table. We fill the table for intention to execution, backtrace to recover the alignment, and see how the same machinery generalizes to weighted edits, Viterbi, and biological sequence alignment.\n",{"path":18106,"title":18107,"module":14054,"summary":18108},"\u002Fnatural-language-processing\u002Ffoundations\u002Fn-gram-language-models","N-Gram Language Models","A language model assigns a probability to a sequence of words and, equivalently, predicts the next word from its history. The n-gram model makes this tractable by truncating the history to the last few words, estimates the resulting conditional probabilities by counting, and is scored by perplexity. We build the model from the chain rule, work a bigram example on a small corpus, and read perplexity as a branching factor. The next lesson covers the zero counts that break this model and the smoothing that repairs them.\n",{"path":18110,"title":18111,"module":14054,"summary":18112},"\u002Fnatural-language-processing\u002Ffoundations\u002Fsmoothing-and-backoff","Smoothing and Backoff","Every finite corpus is missing good word sequences it simply never saw, so a raw n-gram model assigns them probability zero and breaks. Smoothing repairs the zeros: add-one and add-k shave mass off seen events, backoff and interpolation fall back on shorter contexts, and Kneser-Ney — worked here by hand — replaces raw frequency with how many contexts a word completes. We close on web-scale stupid backoff and the neural models that dissolve the zero problem rather than patch it.\n",{"path":18114,"title":18115,"module":18116,"summary":18117},"\u002Fnatural-language-processing\u002Fclassification\u002Fnaive-bayes-and-sentiment","Naive Bayes and Sentiment Classification","Text Classification","Text classification assigns a category to a document — positive or negative, spam or not, one topic among many. Naive Bayes is a generative solution: apply Bayes' rule, assume the words are conditionally independent given the class, and the winning class is the one maximizing the product of a prior and per-word likelihoods. We train it by counting with add-one smoothing, work a full sentiment example by hand, sharpen it for sentiment (binary counts, negation, lexicons), and place it among the transformer classifiers that came after.\n",{"path":18119,"title":18120,"module":18116,"summary":18121},"\u002Fnatural-language-processing\u002Fclassification\u002Fevaluating-classifiers","Evaluating Classifiers","A trained classifier is only useful once we can measure how good it is. We build the confusion matrix, see why accuracy misleads on unbalanced data, and define precision, recall, and the F-measure that balances them. Multi-class tasks need macro- versus micro-averaging; reliable estimates need cross-validation. We close on statistical significance — the paired bootstrap test for whether one system's lead over another is significant.\n",{"path":18123,"title":18124,"module":18116,"summary":18125},"\u002Fnatural-language-processing\u002Fclassification\u002Flogistic-regression","Logistic Regression","Logistic regression is the discriminative counterpart to naive Bayes: instead of modelling how a document is generated, it learns weights that directly separate the classes. We build it from the sigmoid, derive the cross-entropy loss from maximum likelihood, learn the weights by stochastic gradient descent, regularize to curb overfitting, and generalize to many classes with the softmax. The two-class model is already a one-neuron network, so this is the bridge to neural language models.\n",{"path":18127,"title":18128,"module":18116,"summary":18129},"\u002Fnatural-language-processing\u002Fclassification\u002Fsentiment-and-affect-lexicons","Sentiment and Affect Lexicons","A sentiment lexicon is a list of words annotated with the affective meaning they carry — positive or negative, or scores along valence, arousal, and dominance. We fix what \"emotion\" means (basic-emotion versus dimensional models), survey the standard lexicons, and then build lexicons three ways: by human labeling with best-worst scaling, by semi-supervised induction from seed words over an embedding space, and by supervised learning from starred reviews. We close on connotation frames, which record the sentiment a verb implies about each of its arguments.\n",{"path":18131,"title":18132,"module":18133,"summary":18134},"\u002Fnatural-language-processing\u002Fsemantics\u002Fvector-semantics-and-embeddings","Vector Semantics and Embeddings","Semantics","Vector semantics represents a word's meaning as a point in space, derived from the company the word keeps. This first part builds the count-based side: the distributional hypothesis, co-occurrence matrices in their term-document and word-word forms, cosine as the similarity measure, and the two weightings — tf-idf and PPMI — that fix what raw counts get wrong. The result is a sparse, interpretable vector for every word, and the setup for the dense embeddings of the next lesson.\n",{"path":18136,"title":18137,"module":18133,"summary":18138},"\u002Fnatural-language-processing\u002Fsemantics\u002Fstatic-word-embeddings","Static Word Embeddings: word2vec and After","Count-based vectors are long and sparse; embeddings are the short, dense alternative. This lesson builds them with word2vec's skip-gram and negative sampling — a classifier whose learned weights are the vectors — derives its gradient, and works one update by hand. It then reads relations off the analogy parallelogram, surveys the papers that framed the static-embedding era (word2vec, GloVe, the SGNS-as-PPMI equivalence, fastText, ELMo), and closes on the biases embeddings inherit and the single-vector-per-word ceiling that contextual models break.\n",{"path":17,"title":18140,"module":18133,"summary":18141},"Neural Networks and Neural Language Models","A neural network is a stack of units, each a weighted sum passed through a non-linearity — a single unit on its own is logistic regression. We build the network up from that unit: the activation functions that give it power, the XOR problem that forces a hidden layer, the feedforward forward pass in matrix form, and the Bengio-style feedforward neural language model that concatenates word embeddings and predicts the next word with a softmax. Training is cross-entropy minimized by gradient descent, with backpropagation supplying the gradient. Embeddings let the model share statistical strength across similar words, avoiding the sparsity that limits n-gram models.\n",{"path":5656,"title":18143,"module":13116,"summary":18144},"Sequence Labeling: POS and NER","Sequence labeling assigns one tag to every token in a sentence. This first part sets up the task through its two canonical cases — part-of-speech tagging over the Penn Treebank tagset, and named-entity recognition reframed as token labeling with the BIO scheme — then builds the hidden Markov model, the classic probabilistic tagger. The HMM tags by Bayesian inference: transition and emission probabilities under two Markov assumptions, reducing tagging to an argmax over tag sequences. That argmax is exponential to enumerate, which sets up the Viterbi decoder, the CRF, and neural taggers of the next lesson.\n",{"path":18146,"title":18147,"module":13116,"summary":18148},"\u002Fnatural-language-processing\u002Fsequences\u002Fcrfs-and-neural-taggers","Viterbi Decoding, CRFs, and Neural Taggers","The HMM reduced tagging to an argmax over exponentially many tag sequences. This lesson builds the decoder that makes it tractable — the Viterbi dynamic program, worked through a full numeric trace on real WSJ probabilities — then keeps that same decoder while replacing the HMM's rigid tables. The linear-chain conditional random field is a discriminative log-linear model whose global feature functions can inspect any part of the input, which is why CRFs win for NER. Finally it traces the shift to neural taggers (biLSTM-CRF, character-aware NER, ELMo), where hand-built features become learned representations while the Viterbi decoder carries over unchanged.\n",{"path":13118,"title":5,"module":13116,"summary":13135},{"path":11834,"title":18151,"module":15001,"summary":18152},"Transformers and Self-Attention","Recurrence forced language models to read one word at a time and to squeeze every dependency through a chain of hidden states. Self-attention removes the recurrence: at every layer each position compares itself to every other and reads a weighted mixture of them, in a single parallel step. This first part builds the attention operation from the ground up — the soft lookup, queries and keys and values, the scaled dot-product, the numeric trace, the matrix form, and the causal mask — and sets up the full transformer architecture that follows.\n",{"path":18154,"title":16710,"module":15001,"summary":18155},"\u002Fnatural-language-processing\u002Ftransformers\u002Fthe-transformer-architecture","This part takes the scaled dot-product attention of the previous lesson and assembles the full transformer architecture around it: multi-head attention so several relations can be read at once, the transformer block of residual connections and layer norm that makes deep stacks trainable, positional embeddings that restore word order, the decoder-only language model, and the encoder, decoder, and encoder-decoder shapes — closing with the 2017 paper and the pre-norm, FlashAttention, and RoPE refinements that scaled it up.\n",{"path":18157,"title":16818,"module":15001,"summary":18158},"\u002Fnatural-language-processing\u002Ftransformers\u002Flarge-language-models","A large language model is a decoder-only transformer trained on one objective — predict the next token. This first part assembles the inference side: the language-modeling head that turns a hidden state into a distribution over the vocabulary, autoregressive generation, and the decoding strategies — greedy, beam, and sampling with temperature, top-k, and nucleus — that read text back out of that distribution. Training the distribution at web scale comes next.\n",{"path":18160,"title":18161,"module":15001,"summary":18162},"\u002Fnatural-language-processing\u002Ftransformers\u002Fllm-pretraining-and-scaling","Large Language Models: Pretraining and Scaling","A language model's next-token distribution is only as good as the parameters behind it. This part is where those parameters come from: self-supervised pretraining on web-scale text with teacher forcing and cross-entropy, the scaling laws that make test loss a predictable power law in parameters, data, and compute, the KV cache that keeps long-context inference affordable, and how a finished model is evaluated by perplexity and benchmarks — closing with the Kaplan, Chinchilla, GPT-3, and emergence papers behind the scaling story.\n",{"path":18164,"title":18165,"module":15001,"summary":18166},"\u002Fnatural-language-processing\u002Ftransformers\u002Ffine-tuning-and-prompting","Fine-Tuning and Prompting","A pretrained transformer is a general-purpose knowledge source; a task is what you do with it. There are two ways to adapt one, and this first part covers the one that updates the weights: fine-tuning. A bidirectional encoder like BERT is pretrained by masked language modeling, then a small task head is bolted on and the whole thing is trained on labelled data for classification, sequence labeling, or span-based question answering — with parameter-efficient variants (adapters, LoRA) that touch only a sliver of the weights. Prompting, the family that leaves the weights frozen, comes next.\n",{"path":18168,"title":18169,"module":15001,"summary":18170},"\u002Fnatural-language-processing\u002Ftransformers\u002Fprompting-and-alignment","Prompting and Alignment","Fine-tuning adapts a model by changing its weights. The second family of adaptation changes nothing: a large frozen model performs a task from an instruction and a few examples placed in its context. This part covers prompting and in-context learning, chain-of-thought that elicits reasoning, and the two training stages — instruction tuning and RLHF — that turn a fluent base predictor into an aligned assistant, closing with the BERT, LoRA, chain-of-thought, InstructGPT, and retrieval-augmentation papers behind the modern adaptation pipeline.\n",{"path":18172,"title":18173,"module":18174,"summary":18175},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fconstituency-parsing","Constituency Parsing","Linguistic Structure","A constituency parse groups a sentence into nested phrases described by a context-free grammar. We build the CFG formalism, read the phrase structure of English off a treebank, confront the structural ambiguity that makes parsing hard, convert to Chomsky normal form, and then solve it with CKY — the dynamic-programming chart that fills a triangular table bottom-up. Probabilistic and neural span parsers, evaluation, and shallow parsing follow in the companion lesson.\n",{"path":18177,"title":18178,"module":18174,"summary":18179},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcky-scoring-and-evaluation","CKY Scoring, Evaluation, and Shallow Parsing","The CKY chart returns every parse but does not say which is correct. Disambiguation needs a score on trees. This lesson attaches probabilities to a grammar (the PCFG and lexicalization), replaces the grammar with a neural span scorer over a pretrained encoder, states the self-attentive results that made it the state of the art, evaluates parsers against a treebank with PARSEVAL, and closes with chunking and shallow parsing for tasks that need only the flat phrases.\n",{"path":18181,"title":18182,"module":18174,"summary":18183},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fdependency-parsing","Dependency Parsing","A dependency parse throws away phrases and keeps only directed, labeled arcs from heads to their dependents, so the subject and object of a verb hang off the verb directly. We fix the formalism (rooted trees, typed Universal-Dependency relations, projectivity), then build the first parser family: transition-based arc-standard and arc-eager parsing, a greedy stack-and-buffer machine trained from an oracle. Graph-based and neural dependency parsing follow in the companion lesson.\n",{"path":18185,"title":18186,"module":18174,"summary":18187},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fgraph-based-and-neural-dependency-parsing","Graph-Based and Neural Dependency Parsing","Greedy transition parsing commits locally; the graph-based family scores whole trees instead. This lesson scores every candidate head-dependent edge and extracts the maximum spanning tree with Chu-Liu\u002FEdmonds, develops the biaffine neural scorer that made graph-based parsing the accuracy leader, evaluates parsers with the unlabeled and labeled attachment scores (UAS and LAS), and closes on where the two parser families sit and what they feed downstream.\n",{"path":18189,"title":18190,"module":18174,"summary":18191},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fword-senses-and-wsd","Word Senses and Disambiguation","A word is not an atom of meaning: \"bass\" names a fish, a voice, and an instrument, and one static embedding blurs them into a single point. This lesson pulls those senses apart. We define polysemy and the relations that organize senses — synonymy, antonymy, hyponymy, meronymy — build them into WordNet's synset graph, measure similarity along that graph, and then solve the core of word sense disambiguation: the most-frequent-sense baseline, the Lesk gloss-overlap algorithm, feature-based classifiers, and the nearest-neighbor method over BERT embeddings. WSD variants, embeddings, and evaluation follow in the companion lesson.\n",{"path":18193,"title":18194,"module":18174,"summary":18195},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fwsd-in-practice-and-induction","WSD in Practice and Word Sense Induction","Beyond core word sense disambiguation lie the variants and loose ends: the sense-inventory-free Word-in-Context task, retrofitting static embeddings to a thesaurus, discovering senses without a fixed inventory (word sense induction), the gloss-aware and bi-encoder neural systems that hold the state of the art, and how WSD and its cousins are evaluated. Together they connect one-vector-per-word embeddings to sense-aware contextual representations.\n",{"path":18197,"title":18198,"module":18174,"summary":18199},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fsemantic-roles-and-information-extraction","Semantic Roles and Information Extraction","Semantic roles answer \"who did what to whom\" for a single event, abstracting away the syntax that expresses it. We show why syntax alone is not enough, generalize over diathesis alternations with thematic roles, number a predicate's arguments with PropBank and group predicates into frames with FrameNet, tag each argument automatically with semantic role labeling, and factor predicates into primitives. Information extraction scales the idea to a corpus in the companion lesson.\n",{"path":18201,"title":18202,"module":18174,"summary":18203},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Frelations-events-and-templates","Relations, Events, and Templates","Semantic roles answer \"who did what\" for one predicate; information extraction scales the idea to a whole corpus. This lesson turns unstructured text into structured data: relation extraction pulls entity-relation-entity triples out of sentences by patterns, supervision, and distant supervision; event and temporal extraction place those facts on a timeline; and template filling and knowledge-base population assemble them into a database a downstream system can query.\n",{"path":18205,"title":18206,"module":18174,"summary":18207},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcoreference-and-discourse","Coreference and Discourse","A text is more than a bag of sentences: entities recur under different names. Coreference resolution links every mention to the discourse entity it evokes — the linguistic background of pronouns, definite NPs, and names; mention detection; the mention-pair, mention-ranking, and entity-based architectures; a neural end-to-end span model that scores candidate antecedents; features, evaluation by the CoNLL F1, gender bias, and the neural coreference lineage. Discourse coherence follows in the companion lesson.\n",{"path":18209,"title":18210,"module":18174,"summary":18211},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcoherence-and-discourse-structure","Coherence and Discourse Structure","Coherence is what makes a run of sentences a discourse rather than an arbitrary collection. This lesson develops coherence relations and Rhetorical Structure Theory trees, discourse-structure parsing, Centering and the entity grid for entity-based coherence, and representation-learning models of local coherence, measured in part over the coreference chains recovered in the companion lesson.\n",{"path":18213,"title":18214,"module":18174,"summary":18215},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Flogical-semantics","Logical Representations of Meaning","A meaning representation turns a sentence into a formal structure a machine can check against a world and reason over. We set the desiderata a good representation must meet, ground truth in a model, build up first-order logic for sentences with its connectives, quantifiers, and inference, and reify events with the neo-Davidsonian event variable to escape fixed predicate arity. The compositional lambda calculus, quantifier scope, and description logics follow in the companion lesson.\n",{"path":18217,"title":18218,"module":18174,"summary":18219},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fcompositional-semantics-and-description-logics","Compositional Semantics and Description Logics","How do you compute a logical form from a sentence automatically? This lesson builds the compositional machinery: the lambda calculus that assembles a formula from a parse tree one beta-reduction at a time, the quantifier-scope ambiguity a single syntax tree leaves open, and the decidable description logics — TBox, ABox, subsumption, role restrictions — behind the Web Ontology Language, closing with how the map from string to logical form can be learned.\n",{"path":18221,"title":18222,"module":18174,"summary":18223},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fsemantic-parsing","Semantic Parsing","Turning a sentence into a structured, executable meaning, the grammar-based way. We take the logical forms defined earlier and build them compositionally: a rule-based parser that walks a syntax tree applying lambda terms, then Combinatory Categorial Grammar (CCG), which fuses syntax and semantics so one lexicalized derivation produces both — including supertagging and A* parsing. Learned and neural semantic parsers follow in the companion lesson.\n",{"path":18225,"title":18226,"module":18174,"summary":18227},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fneural-semantic-parsing","Learned and Neural Semantic Parsing","Hand-writing a lexicon of lambda terms does not scale, so this lesson learns the parser instead. We cover the two supervision regimes (from logical forms and from denotations), Abstract Meaning Representation as a rooted concept graph, neural sequence-to-sequence parsing with constrained decoding and copy mechanisms, executable text-to-SQL and knowledge-based question answering, the practical systems that made learned parsers accurate, and how the task is evaluated.\n",{"path":18229,"title":18230,"module":18174,"summary":18231},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Finformation-extraction","Information Extraction","Information extraction turns free text into a database, and the first step is relation extraction: pulling entity-relation-entity triples out of sentences. We cover all five families — hand-built patterns, supervised classifiers, semi-supervised bootstrapping, distant supervision, and unsupervised Open IE — with worked bootstrapping and distant-supervision traces, then the neural and LLM systems that extended them. Times, events, and templates follow in the companion lesson.\n",{"path":18233,"title":18234,"module":18174,"summary":18235},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Ftimes-events-and-templates","Extracting Times, Events, and Templates","Once relation extraction has produced typed triples, the information-extraction pipeline still has to place facts in time and assemble them into records. This lesson detects and normalizes temporal expressions to ISO 8601 values, detects events and orders them on a timeline with the 13 Allen relations, and fills slot-and-filler templates — flat and hierarchical — for stereotyped situations, closing the loop from text to a queryable database.\n",{"path":18237,"title":18238,"module":18174,"summary":18239},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fdiscourse-coherence","Discourse Coherence","A text is more than a set of sentences. What binds a run of sentences into a discourse is coherence, and one of its sources is structured relations between clauses. This lesson develops relational coherence — RST and the PDTB models of coherence relations — and discourse-structure parsing: EDU segmentation and shift-reduce RST parsing, then PDTB relation classification. Entity-based and global coherence follow in the companion lesson.\n",{"path":18241,"title":18242,"module":18174,"summary":18243},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fentity-based-and-global-coherence","Entity-Based and Global Coherence","A text coheres not only through relations between clauses but by staying about the same entities and the same topic, and by obeying the macro-structure of its genre. This lesson develops Centering Theory and the entity grid for entity-based coherence, representation-learning models of local coherence, and global coherence — topic segmentation, narrative and argumentation structure, and scientific discourse — then the neural models that learn each.\n",{"path":18245,"title":18246,"module":18174,"summary":18247},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Fconstituency-grammars","Constituency Grammars","A constituency grammar is the declarative theory of sentence structure that a parser operates on. We build the context-free grammar formalism from its four parts, show how derivations become parse trees, and work through the phrase structure of English — noun phrases, verb phrases and their subcategorization frames, agreement, coordination, and long-distance dependencies. The treebank, normal-form, and lexicalized views follow in the companion lesson.\n",{"path":18249,"title":18250,"module":18174,"summary":18251},"\u002Fnatural-language-processing\u002Flinguistic-structure\u002Ftreebanks-and-lexicalized-grammars","Treebanks and Lexicalized Grammars","Where does a grammar come from, and how is it prepared for a parser? We read a context-free grammar off the Penn Treebank, normalize it to Chomsky Normal Form for the CKY chart, then invert the phrase-structure emphasis with lexicalized grammars — Combinatory Categorial Grammar and its slash categories — and close with the grammar's fate in the neural era: span scoring, self-attention, and grammar induction.\n",{"path":18253,"title":18254,"module":16806,"summary":18255},"\u002Fnatural-language-processing\u002Fapplications\u002Fmachine-translation","Machine Translation","Machine translation is the task that built the modern toolkit: the encoder-decoder was invented for it, attention was invented to fix its fixed-context bottleneck, and both were later folded into the general transformer. We work through why translation is hard (word order, morphology, lexical and structural divergences), the sequence-to-sequence model and its attention mechanism, transformer-based NMT with cross-attention, subword tokenization with a shared vocabulary, beam-search decoding, and evaluation by BLEU and its successors chrF, BERTScore, and COMET — closing on multilingual and low-resource translation and backtranslation.\n",{"path":18257,"title":18258,"module":16806,"summary":18259},"\u002Fnatural-language-processing\u002Fapplications\u002Fmachine-translation-decoding-and-evaluation","Machine Translation: Decoding, Evaluation, and Scale","Having built the transformer translation model, we now decode from it and measure the output. Beam search turns the decoder's per-step distributions into a single output string; length normalization keeps it from favoring short translations. We then score translations automatically — BLEU with its n-gram precision, clipping, and brevity penalty, worked through by hand, then its successors chrF, BERTScore, and COMET — and close on the parts of MT that scale beyond one language pair: multilingual and low-resource translation, backtranslation, gender bias, and the lineage from the Transformer to massively multilingual models like NLLB-200.\n",{"path":18261,"title":18262,"module":16806,"summary":18263},"\u002Fnatural-language-processing\u002Fapplications\u002Fquestion-answering","Question Answering","A question-answering system takes a natural-language question and returns an answer, not a ranked list of documents. Almost every modern system is built on one pattern: retrieve then read. We start with the information-retrieval machinery that finds candidate text — tf-idf and BM25 term weighting, a worked ranking example, the inverted index, and dense embedding retrieval — then build the retriever-reader pipeline that extracts an answer span with BERT and trace a full retrieve-and-read example end to end.\n",{"path":18265,"title":18266,"module":16806,"summary":18267},"\u002Fnatural-language-processing\u002Fapplications\u002Fquestion-answering-knowledge-and-llms","Question Answering: Knowledge Bases and Language Models","The retrieve-and-read pipeline extracts an answer span from prose, but not all knowledge lives in prose. This part covers the rest of the QA stack: entity linking (Wikification) that grounds a question's entities to a knowledge base, knowledge-based QA by semantic parsing a question into an executable query, and the modern default — closed-book QA and retrieval-augmented generation with a large language model — closing on the DPR\u002FRAG\u002Ffusion-in-decoder lineage and how factoid answers are scored by exact match and F1.\n",{"path":18269,"title":18270,"module":16806,"summary":18271},"\u002Fnatural-language-processing\u002Fapplications\u002Fdialogue-and-chatbots","Dialogue and Chatbots","Conversation is the most natural interface to a machine and one of the hardest to build. We set up what makes human dialogue work — turns, speech acts, grounding, and the local structure of adjacency pairs — then trace the two traditions that answer it: chatbots built to chat (ELIZA's pattern-matching, corpus retrieval, and seq2seq generation with its blandness problem) and task-oriented systems built to get something done (the GUS frame-and-slot architecture and the modern NLU \u002F state-tracker \u002F policy \u002F NLG pipeline that accumulates a frame across turns).\n",{"path":18273,"title":18274,"module":16806,"summary":18275},"\u002Fnatural-language-processing\u002Fapplications\u002Fdialogue-systems-and-assistants","Dialogue Systems: LLM Assistants, Evaluation, and Design","Two dialogue traditions — chatbots built to chat and task-oriented frame systems built to get something done — met in the aligned LLM assistant. Instruction tuning plus RLHF fold chit-chat and task dialogue into one model; the LaMDA \u002F InstructGPT \u002F ChatGPT lineage fills in how. The lesson then turns to evaluation (human ratings and acute-eval for chatbots, task success and slot error rate for task systems), user-centered design with Wizard-of-Oz prototyping, and the ethical stakes of building agents people talk to.\n",{"path":18277,"title":18278,"module":16806,"summary":18279},"\u002Fnatural-language-processing\u002Fapplications\u002Ftext-summarization","Text Summarization","Summarization compresses a document to its essential meaning, by either selecting sentences to keep (extractive) or writing new ones (abstractive). This part fixes the task and its flavors — single vs. multi-document, generic vs. query-focused, extractive vs. abstractive — then works through extractive summarization in full: scoring by position and centrality, the TextRank\u002FLexRank graph algorithm run as PageRank over a sentence-similarity graph with a worked iteration, and supervised sentence selection.\n",{"path":18281,"title":18282,"module":16806,"summary":18283},"\u002Fnatural-language-processing\u002Fapplications\u002Fabstractive-summarization-and-evaluation","Abstractive Summarization and Evaluation","Extractive methods can only reuse the source's own sentences; to compress within a sentence or paraphrase, a summarizer has to generate. This part covers abstractive summarization: the sequence-to-sequence approach, the pointer-generator's copy switch and coverage mechanism, pretrained summarizers (BART, PEGASUS) and zero-shot LLM prompting, the long-document and factuality problems, and ROUGE evaluation with a worked example and its limits — closing on the abstractive lineage from See 2017 through faithfulness metrics.\n",{"path":18285,"title":18286,"module":18287,"summary":18288},"\u002Fnatural-language-processing\u002Fspeech\u002Fphonetics","Phonetics","Speech","Before a recognizer can read speech it has to know what speech is. This first part covers the linguistic substrate: phones and their transcription in the IPA and ARPAbet; articulatory phonetics — how the vocal tract shapes airflow into consonants and vowels; and prosody — stress, tune, and the F0 contour. The acoustic side — the waveform, its spectrum, formants, and the spectrogram — is the second part.\n",{"path":18290,"title":18291,"module":18287,"summary":18292},"\u002Fnatural-language-processing\u002Fspeech\u002Facoustic-phonetics","Acoustic Phonetics","Articulation is the cause; the acoustic signal is the effect, and the effect is all a microphone ever gets. This part follows the sound out of the mouth: waves, sampling and the Nyquist limit, F0 and the pitch track, the mel scale, the spectrum and Fourier analysis, the source-filter model that explains why each vowel carries its own formants, and the spectrogram the log-mel front end of every ASR system sits directly on top of — closing with neural TTS, wav2vec, HuBERT, and Whisper, where phonetics went in neural speech.\n",{"path":18294,"title":18295,"module":18287,"summary":18296},"\u002Fnatural-language-processing\u002Fspeech\u002Fautomatic-speech-recognition","Automatic Speech Recognition","Speech recognition maps an acoustic waveform to a string of words, and once the waveform is turned into a sequence of log-mel spectrogram frames the problem is the same sequence-to-sequence transduction the rest of the course already solved. This first part builds the feature front end (framing, the DFT, the mel filterbank, the log), then the modern architectures: the attention-based encoder-decoder, the CTC alignment trick that collapses repeated and blank frames, and RNN-T for streaming. Training-data advances, evaluation, TTS, and the other speech tasks come next.\n",{"path":18298,"title":18299,"module":18287,"summary":18300},"\u002Fnatural-language-processing\u002Fspeech\u002Fasr-evaluation-and-applications","ASR Evaluation and Speech Applications","A recognizer turns a waveform into text; this part scores that text and puts the same machinery to other uses. It opens with the self-supervised and weakly- supervised systems (wav2vec 2.0, HuBERT, Whisper) that made ASR error rates fall. Word error rate reuses the edit distance from the first module, run over words. Text-to-speech runs the whole pipeline in reverse — text to mel spectrogram to waveform. And a family of smaller tasks — wake-word detection, speaker recognition and diarization, language identification — reuse the same log-mel front end without the decoder.\n",{"path":18302,"title":18303,"module":6,"summary":6},"\u002Fnatural-language-processing","Natural Language Processing",{"path":18305,"title":18306,"module":14054,"summary":18307},"\u002Fparticle-physics\u002Ffoundations\u002Fhistorical-overview-particle-zoo","From the Electron to the Particle Zoo","A timeline of the subject, from J. J. Thomson's electron in 1897 to the Higgs boson in 2012. The electron, photon, nucleus, proton, and neutron gave a tidy picture that Yukawa's meson prediction and the muon–pion confusion complicated; strange particles in cosmic rays and the accelerator-era flood of hadrons then produced a \"particle zoo\" that only the quark model organized.\n",{"path":18309,"title":18310,"module":14054,"summary":18311},"\u002Fparticle-physics\u002Ffoundations\u002Fparticle-physics-basic-concepts","Basic Concepts and Particle Classification","Every particle has an antiparticle of equal mass and opposite charge, a consequence of the Dirac equation confirmed by the positron. Feynman diagrams track interactions in spacetime; the material particles sort into leptons and the composite hadrons built from quarks, with baryons carrying three quarks and mesons a quark-antiquark pair.\n",{"path":18313,"title":18314,"module":14054,"summary":18315},"\u002Fparticle-physics\u002Ffoundations\u002Ffundamental-interactions-force-carriers","Fundamental Interactions and Force Carriers","Four interactions account for every force in nature: strong, electromagnetic, weak, and gravitational, in decreasing strength. Each is carried by a boson — the gluon, photon, W and Z, and the graviton — with a range fixed by the carrier's mass through the Yukawa relation, and a coupling constant that itself varies with distance.\n",{"path":18317,"title":18318,"module":18319,"summary":18320},"\u002Fparticle-physics\u002Funits-kinematics\u002Fnatural-units-and-scales","Natural Units and Scales","Units and Kinematics","Setting $\\hbar = c = 1$ collapses mass, momentum, and energy into a single unit, the GeV, and turns lengths and times into inverse energies through the conversion $\\hbar c = 197.3$ MeV·fm. This lesson fixes the natural-unit conventions used for the rest of the course, converts cross sections between barns and GeV$^{-2}$, and shows how to restore factors of $\\hbar$ and $c$ by dimensional analysis.\n",{"path":18322,"title":18323,"module":18319,"summary":18324},"\u002Fparticle-physics\u002Funits-kinematics\u002Ffour-vectors-invariant-mass","Four-Vectors and Invariant Mass","The energy and momentum of a particle form a four-vector whose square is the frame-independent quantity $p^2 = m^2$. This lesson develops the metric and four-vector products, the invariant mass of a multiparticle system, the center-of-momentum and laboratory frames, and the description of collinear boosts by rapidity, whose additivity replaces the awkward velocity-addition law.\n",{"path":18326,"title":18327,"module":18319,"summary":18328},"\u002Fparticle-physics\u002Funits-kinematics\u002Fdecay-scattering-kinematics-mandelstam","Decay, Scattering, and Mandelstam Variables","Two-body decay in the rest frame fixes the daughter momenta from the three masses alone; production thresholds follow from the minimum invariant mass. This lesson works both, then introduces the Mandelstam invariants $s$, $t$, $u$ for $2\\to2$ scattering, proves the identity $s+t+u=\\sum m_i^2$, and maps the physical regions and the crossing that relates channels.\n",{"path":18330,"title":18331,"module":18319,"summary":18332},"\u002Fparticle-physics\u002Funits-kinematics\u002Fcross-sections-golden-rule","Cross Sections and the Golden Rule","The cross section measures how often a scattering happens and the decay width how fast a particle disintegrates. This lesson defines both, relates event rate to luminosity through $R=\\mathcal L\\,\\sigma$ and lifetime to width through $\\tau=\\hbar\u002F\\Gamma$, and states Fermi's golden rule with Lorentz-invariant phase space, giving the master formulas that turn an amplitude $\\mathcal M$ into a measurable rate for $1\\to2$ decay and $2\\to2$ scattering.\n",{"path":18334,"title":18335,"module":18336,"summary":18337},"\u002Fparticle-physics\u002Fsymmetries\u002Fconservation-laws-symmetries","Conservation Laws and Symmetries","Symmetries and Conservation Laws","Which decays occur is decided by conservation laws, each tied by Noether's theorem to a symmetry of physical law. Energy, charge, baryon number, and lepton number are conserved universally; strangeness, isospin, and parity hold in the strong and electromagnetic interactions but break in the weak one, whose parity and CP violation distinguish matter from antimatter.\n",{"path":18339,"title":18340,"module":18336,"summary":18341},"\u002Fparticle-physics\u002Fsymmetries\u002Fdiscrete-symmetries-cpt","Discrete Symmetries — C, P, T, and CPT","Parity reflects space, charge conjugation swaps particle for antiparticle, and time reversal runs the clock backward. Each assigns multiplicative quantum numbers that act as selection rules — intrinsic parities, the photon's C = −1, the C-parity argument fixing the pion's two-photon decay. Their product CPT is a theorem of any local relativistic field theory, forcing particle and antiparticle to share mass and lifetime.\n",{"path":18343,"title":18344,"module":18336,"summary":18345},"\u002Fparticle-physics\u002Fsymmetries\u002Fparity-violation-weak","Parity Violation and the Weak Force","The tau–theta puzzle forced a choice: two particles with identical mass but opposite parity, or one particle whose decay violates parity. Lee and Yang proposed the latter, Wu's polarized cobalt-60 confirmed it, and the violation proved maximal. The charged weak force couples only to left-handed chirality — the Goldhaber experiment showed the neutrino is left-handed — which is why the mirror image of a weak decay is something nature never produces.\n",{"path":18347,"title":18348,"module":18336,"summary":18349},"\u002Fparticle-physics\u002Fsymmetries\u002Fsu2-su3-flavor-symmetry","Isospin, SU(2), and Flavor SU(3)","The near-equal masses of the proton and neutron, and of the three pions, signal a continuous internal symmetry of the strong force: isospin, an SU(2) whose ladder operators move between the members of a multiplet. Adding strangeness enlarges it to an approximate SU(3) flavor symmetry, and the Gell-Mann–Nishijima relation Q = I3 + Y\u002F2 places every hadron on a weight diagram in the isospin–hypercharge plane — the language in which the quark model is written.\n",{"path":18351,"title":18352,"module":18353,"summary":18354},"\u002Fparticle-physics\u002Fquark-model\u002Feightfold-way-su3","The Eightfold Way and SU(3) Flavor","The Quark Model","Gell-Mann and Ne'eman's classification of the hadrons into geometric multiplets, read as representations of an approximate flavor SU(3). The fundamental triplet (u, d, s) and its antitriplet combine into the meson nonet from 3⊗3̄ = 8⊕1 and the baryon octet and decuplet from 3⊗3⊗3, and the empty corner of the decuplet forecast the Ω⁻.\n",{"path":18356,"title":18357,"module":18353,"summary":18358},"\u002Fparticle-physics\u002Fquark-model\u002Fmeson-spectroscopy","Meson Multiplets and Quantum Numbers","Mesons as quark–antiquark bound states. The spin singlet and triplet, orbital excitations, and the assignment of J^PC from the quark spins and orbital angular momentum, giving the pseudoscalar and vector nonets. The η–η' and ω–φ mixing problems, and the charmonium and bottomonium spectra read as heavy-quark positronium.\n",{"path":18360,"title":18361,"module":18353,"summary":18362},"\u002Fparticle-physics\u002Fquark-model\u002Fbaryon-spectroscopy","Baryon Multiplets, Spin, and the Color Puzzle","Baryons as three-quark states, with a wavefunction factored into space, spin, flavor, and color. The spin-3\u002F2 Δ⁺⁺ = uuu forces a totally symmetric state that the Pauli principle forbids, and the resolution is an antisymmetric color factor — the first evidence for color. The octet and decuplet spin content, and baryon magnetic moments as a quantitative test of the model.\n",{"path":18364,"title":18365,"module":18353,"summary":18366},"\u002Fparticle-physics\u002Fquark-model\u002Fcolor-confinement-exotics","Color, Confinement, and Exotic Hadrons","Color as the gauged SU(3) charge, and the requirement that every physical hadron be a color singlet — which selects q-qbar mesons and qqq baryons as the simplest states. The R-ratio of e⁺e⁻ annihilation measures three colors directly. Beyond the simplest singlets lie glueballs, tetraquarks, and pentaquarks, and the recent XYZ states, read as either compact multiquarks or loose hadronic molecules.\n",{"path":18368,"title":18369,"module":18370,"summary":18371},"\u002Fparticle-physics\u002Frelativistic-wave-equations\u002Fklein-gordon-equation","The Klein-Gordon Equation","Relativistic Wave Equations","Quantizing the relativistic energy relation $E^2 = p^2 + m^2$ produces the Klein-Gordon equation for a scalar field. Its plane-wave solutions come in positive- and negative-energy branches, and the conserved density it supplies is not positive-definite — the two difficulties that first drove physicists to seek a first-order equation. The static Klein-Gordon equation with a point source gives the Yukawa potential, and the free equation gives the scalar propagator that later modules attach to exchanged lines.\n",{"path":18373,"title":18374,"module":18370,"summary":18375},"\u002Fparticle-physics\u002Frelativistic-wave-equations\u002Fdirac-equation-spinors","The Dirac Equation and Spinors","Dirac demanded a wave equation first order in time to fix the Klein-Gordon density problem. Factorizing $E^2 = p^2 + m^2$ into a linear form forces the coefficients to be anticommuting matrices — the gamma matrices of the Clifford algebra — so the wavefunction becomes a four-component spinor. The plane-wave solutions split into two particle and two antiparticle states, spin appears automatically with the correct $g = 2$ magnetic moment, and the chirality projectors that the weak interaction later needs fall straight out of the fifth gamma matrix.\n",{"path":18377,"title":18378,"module":18370,"summary":18379},"\u002Fparticle-physics\u002Frelativistic-wave-equations\u002Fantiparticles-hole-theory","Antiparticles and Hole Theory","The negative-energy solutions of the Dirac equation refuse to go away, so they must mean something. Dirac read them as a filled sea of occupied negative-energy states whose holes are positive-energy antiparticles, predicting the positron before its discovery. The picture works for fermions but not bosons, and the Feynman-Stückelberg interpretation replaces it: an antiparticle is a negative-energy solution propagating backward in time, equivalent to a positive-energy antiparticle going forward. Crossing symmetry ties incoming particles to outgoing antiparticles in a single amplitude.\n",{"path":18381,"title":18382,"module":18383,"summary":18384},"\u002Fparticle-physics\u002Fqed\u002Ffeynman-rules-qed","Feynman Rules for QED","Quantum Electrodynamics","Quantum electrodynamics computes a process by summing diagrams, each a term in a power series in the coupling. Every diagram translates into an amplitude by a fixed dictionary: spinors and polarization vectors for external lines, propagators for internal lines, and the vertex factor $ie\\gamma^\\mu$ for each photon-fermion junction. Squaring the amplitude and feeding it to Fermi's golden rule produces a cross section or decay rate, with each extra vertex costing one power of $\\alpha$.\n",{"path":18386,"title":18387,"module":18383,"summary":18388},"\u002Fparticle-physics\u002Fqed\u002Fqed-tree-processes","Tree-Level QED Processes","The Feynman rules become numbers on the reference reactions of QED. Muon pair production $e^+e^-\\to\\mu^+\\mu^-$ sets the scale with its $1+\\cos^2\\theta$ distribution and $4\\pi\\alpha^2\u002F3s$ total cross section, and its ratio to hadron production counts colors. Compton scattering gives the Klein-Nishina formula and the Thomson limit; Bhabha scattering shows $s$- and $t$-channel interference. Casimir's trick turns every spin-averaged square into a trace of gamma matrices.\n",{"path":18390,"title":18391,"module":18383,"summary":18392},"\u002Fparticle-physics\u002Fqed\u002Frenormalization-running-coupling","Renormalization and the Running Coupling","Beyond tree level, QED loops diverge. The three primitive one-loop diagrams — vacuum polarization, electron self-energy, and vertex correction — carry ultraviolet divergences that regularization exposes as logarithms of a cutoff. Renormalization absorbs them into the measured mass, charge, and field normalization, leaving finite predictions. The surviving physical content is that the coupling runs: vacuum polarization screens charge, so $\\alpha$ grows from $1\u002F137$ at low energy to about $1\u002F128$ at the $Z$ mass.\n",{"path":18394,"title":18395,"module":18383,"summary":18396},"\u002Fparticle-physics\u002Fqed\u002Felectron-g-2","The Anomalous Magnetic Moment","The Dirac equation predicts $g=2$; loops shift it. Schwinger's one-loop vertex correction gives the anomaly $a=(g-2)\u002F2=\\alpha\u002F2\\pi$, and the QED series continues to five loops. The electron $a_e$ agrees with theory to better than a part in a billion, the most precise confrontation of theory and experiment in physics. The muon $a_\\mu$, heavier and so more sensitive to virtual heavy states, is dominated by hadronic uncertainty and sits at the center of a long-running comparison with the Standard Model prediction.\n",{"path":18398,"title":18399,"module":18400,"summary":18401},"\u002Fparticle-physics\u002Fweak-interaction\u002Fva-structure-weak","The V–A Charged Weak Current","The Weak Interaction","Fermi modelled beta decay as a four-fermion contact interaction, but a coupling with dimensions of inverse mass squared makes cross sections grow without bound and the theory fails near 300 GeV. The cure is a heavy mediator: the $W$ boson, whose propagator collapses to Fermi's contact term at low energy and fixes $G_F\u002F\\sqrt2 = g^2\u002F8M_W^2$. Parity violation dictates the current's form — vector minus axial-vector, coupling only to left-chiral fields — and universality of the coupling ties muon decay, beta decay, and pion decay to one constant. Pion decay's helicity suppression of the electron channel is the sharpest test.\n",{"path":18403,"title":18404,"module":18400,"summary":18405},"\u002Fparticle-physics\u002Fweak-interaction\u002Fw-z-bosons-decays","The W and Z Bosons","The contact theory hides a massive mediator. The charged $W^\\pm$ carries the current that changes flavour; the neutral $Z^0$ carries a current that does not. Both were found at CERN's proton–antiproton collider in 1983 at the masses the electroweak theory demanded. Their decay widths partition into leptonic and hadronic channels, and the $Z$ carries a decisive extra: an invisible width from decays to neutrinos that counts the number of light generations at exactly three. Beta decay and muon decay are re-read at the parton level as $W$ exchange.\n",{"path":18407,"title":18408,"module":18400,"summary":18409},"\u002Fparticle-physics\u002Fweak-interaction\u002Fckm-matrix","Quark Mixing and the CKM Matrix","The quark eigenstates the weak force acts on are not the mass eigenstates. Cabibbo captured this with one rotation angle; the GIM mechanism added a fourth quark to cancel dangerous flavour-changing neutral currents and predicted charm before its discovery. Three generations promote the rotation to the unitary Cabibbo–Kobayashi–Maskawa matrix — three angles and one irreducible complex phase, the sole source of Standard-Model CP violation. The Wolfenstein parametrization exposes its steep hierarchy, and unitarity closes into a triangle whose area measures the phase.\n",{"path":18411,"title":18412,"module":18400,"summary":18413},"\u002Fparticle-physics\u002Fweak-interaction\u002Fcp-violation-kaons-b-mesons","CP Violation in Kaons and B Mesons","The neutral kaon is its own laboratory for CP. Weak box diagrams mix $K^0$ and its antiparticle into short- and long-lived states that should be pure CP eigenstates decaying to two and three pions. In 1964 Cronin and Fitch caught the long-lived kaon decaying to two pions — CP is violated, at the two-per-mille level of $\\epsilon$. Direct violation ($\\epsilon'$) followed, and the $B$ factories turned the CKM phase into a large, clean time-dependent asymmetry measuring $\\sin 2\\beta$. The effect is real but far too small to explain why the universe is made of matter.\n",{"path":18415,"title":18416,"module":18417,"summary":18418},"\u002Fparticle-physics\u002Fqcd\u002Fcolor-su3-gluons","Color SU(3), Gluons, and the QCD Lagrangian","Quantum Chromodynamics","Color is the exact gauged SU(3) charge of the strong force. Gauging it forces eight massless gluons in the adjoint representation and, because the gauge group is non-abelian, three- and four-gluon self-couplings absent from QED. This lesson builds the QCD Lagrangian from the covariant derivative and the non-abelian field strength, states the Feynman rules with their color factors, and computes the Casimir invariants that set the strength of quark-gluon and gluon-gluon coupling.\n",{"path":18420,"title":18421,"module":18417,"summary":18422},"\u002Fparticle-physics\u002Fqcd\u002Fasymptotic-freedom-confinement","Asymptotic Freedom and Confinement","The QCD beta function is negative: gluon self-interaction antiscreens color, so the coupling weakens at short distance (asymptotic freedom) and strengthens at long distance (confinement). This lesson computes the one-loop beta coefficient, solves for the running of alpha_s and the emergent scale Lambda_QCD, and reads the strong-coupling regime as the linear quark-antiquark potential of a color flux tube that breaks by pair creation.\n",{"path":18424,"title":18425,"module":18417,"summary":18426},"\u002Fparticle-physics\u002Fqcd\u002Fdeep-inelastic-scattering-partons","Deep Inelastic Scattering and the Parton Model","Scattering electrons hard off a proton resolves pointlike constituents. This lesson sets up the deep-inelastic kinematics, defines the structure functions F1 and F2, and reads Bjorken scaling as the signature of free spin-half partons. The Callan-Gross relation fixes the parton spin, the structure function becomes a charge-weighted sum of parton distributions, and the slow logarithmic scaling violations expose the gluon through DGLAP evolution.\n",{"path":18428,"title":18429,"module":18417,"summary":18430},"\u002Fparticle-physics\u002Fqcd\u002Fjets-hadronization","Jets, Hadronization, and Testing QCD","Quarks and gluons produced in a collision fragment into collimated sprays of hadrons — jets — whose directions track the underlying partons. This lesson reads two-jet events as the quark and antiquark of electron-positron annihilation, three-jet events as direct evidence of the radiated gluon, and the hadronization step as the flux tube breaking into color singlets. Jet algorithms and event-shape variables turn the pattern into precision measurements of alpha_s.\n",{"path":18432,"title":18433,"module":18434,"summary":18435},"\u002Fparticle-physics\u002Felectroweak-higgs\u002Felectroweak-su2-u1","The Electroweak Theory","Electroweak Unification and the Higgs","The electromagnetic and weak interactions are two faces of a single gauge theory built on $SU(2)_L \\times U(1)_Y$. Left-handed fermions sit in weak-isospin doublets and right-handed fermions in singlets, each carrying a hypercharge fixed by the Gell-Mann–Nishijima relation $Q = T_3 + Y\u002F2$. The four gauge fields $W^{1,2,3}$ and $B$ mix: the charged combinations $W^\\pm$ mediate the charged current, while $W^3$ and $B$ rotate through the Weinberg angle into the massless photon and the massive $Z$. The single angle $\\theta_W$ ties the couplings, the boson masses, and the neutral-current strengths together.\n",{"path":18437,"title":18438,"module":18434,"summary":18439},"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fspontaneous-symmetry-breaking","Spontaneous Symmetry Breaking","A symmetry of the Lagrangian need not be a symmetry of the ground state. When the lowest-energy configuration sits away from the symmetric point, the symmetry is spontaneously broken and the vacuum is one of a degenerate family. Breaking a continuous global symmetry produces one massless scalar — a Goldstone boson — for every broken generator, the flat direction along the vacuum manifold. The Mexican-hat potential and the ferromagnet below its Curie point are the working pictures.\n",{"path":18441,"title":18442,"module":18434,"summary":18443},"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fhiggs-mechanism","The Higgs Mechanism","Gauging a spontaneously broken symmetry converts the would-be Goldstone bosons into the longitudinal polarizations of the gauge fields, which thereby acquire mass. Applied to $SU(2)_L \\times U(1)_Y$ with a single Higgs doublet, three of the four scalar degrees of freedom are eaten by the $W^\\pm$ and $Z$; the fourth survives as the physical Higgs boson, and the photon stays massless. Fermion masses come from Yukawa couplings to the same field, each mass proportional to its coupling times the vacuum expectation value $v \\approx 246$ GeV.\n",{"path":18445,"title":18446,"module":18434,"summary":18447},"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fhiggs-boson-discovery","The Higgs Boson","The Higgs boson is produced at the LHC chiefly through gluon fusion, with vector-boson fusion and associated production as cleaner but rarer channels. It decays most often to $b\\bar b$ and $WW^\\ast$, but the discovery rested on two rare clean modes, $H \\to \\gamma\\gamma$ and $H \\to ZZ^\\ast \\to 4\\ell$, whose narrow invariant-mass peaks emerged over smooth backgrounds. ATLAS and CMS announced a boson near 125 GeV in 2012; its measured spin-parity $0^+$ and its couplings, which scale with particle mass, identify it as the Standard Model Higgs.\n",{"path":18449,"title":18450,"module":18434,"summary":18451},"\u002Fparticle-physics\u002Felectroweak-higgs\u002Fstandard-model","The Standard Model","The Standard Model combines the quark model, quantum chromodynamics, and the electroweak theory. SU(3) symmetry sorts the hadrons and predicted the omega; color explains why only colorless quark combinations exist; QCD gives asymptotic freedom and confinement; and spontaneous symmetry breaking through the Higgs field gives the weak bosons their mass.\n",{"path":18453,"title":18454,"module":18455,"summary":18456},"\u002Fparticle-physics\u002Fneutrinos\u002Fneutrino-oscillations","Neutrino Oscillations","Neutrino Physics","Neutrinos are produced and detected in flavour states, but they propagate as mass states, and the two bases are misaligned. A flavour therefore evolves coherently into a superposition of other flavours with a probability set by the mass-squared splitting and the ratio L\u002FE. This lesson derives the two-flavour oscillation formula, applies it to the solar and atmospheric neutrino deficits, shows how the SNO neutral-current measurement resolved the solar problem, and works out the MSW resonance that amplifies mixing inside the Sun.\n",{"path":18458,"title":18459,"module":18455,"summary":18460},"\u002Fparticle-physics\u002Fneutrinos\u002Fneutrino-mass-pmns","Neutrino Mass and the PMNS Matrix","Three-flavour mixing promotes the single oscillation angle to the unitary Pontecorvo–Maki–Nakagawa–Sakata matrix, parametrised by three angles and a Dirac CP phase. This lesson decomposes the PMNS matrix into three rotations, records the measured angles and mass-squared splittings, lays out the normal and inverted mass orderings, contrasts the large leptonic mixing with the near-diagonal CKM matrix, and collects the absolute-mass bounds from beta decay and cosmology.\n",{"path":18462,"title":18463,"module":18455,"summary":18464},"\u002Fparticle-physics\u002Fneutrinos\u002Fdirac-majorana-experiments","Dirac, Majorana, and Neutrino Experiments","A neutral fermion can carry a mass term forbidden to every charged particle, so the neutrino may be its own antiparticle. This lesson contrasts the Dirac and Majorana mass terms and their state content, derives the seesaw mechanism that ties a tiny light mass to a heavy right-handed partner, presents neutrinoless double-beta decay as the decisive lepton-number test, surveys the reactor, accelerator, solar, and atmospheric sources on a baseline–energy map, and explains why neutrino mass is physics beyond the original Standard Model.\n",{"path":18466,"title":18467,"module":18468,"summary":18469},"\u002Fparticle-physics\u002Fexperiment\u002Faccelerators-luminosity","Accelerators, Colliders, and Luminosity","Accelerators and Detectors","Fixed-target machines waste energy in the center-of-mass motion of the whole system, so the reachable $\\sqrt s$ grows only as the square root of the beam energy, while colliders put the full beam energy into the collision. Circular electron machines are limited by synchrotron radiation scaling as $E^4\u002Fm^4R$; proton machines are limited by bending fields. Luminosity, set by beam current and focusing, converts a cross section into an event rate through $R=\\mathcal L\\,\\sigma$, and integrated luminosity sets the total event count.\n",{"path":18471,"title":18472,"module":18468,"summary":18473},"\u002Fparticle-physics\u002Fexperiment\u002Fdetectors-subsystems","Particle Detectors and Subsystems","A detector reads a collision by the energy particles deposit as they cross matter. Charged particles ionize at the Bethe-Bloch rate, radiate in the field of nuclei above a critical energy, and emit Cherenkov light above a velocity threshold; electrons and photons build electromagnetic showers over a radiation length, and hadrons build wider showers over a nuclear interaction length. The onion of tracker, electromagnetic and hadronic calorimeters, and outer muon chambers turns these processes into momentum, energy, and identity, with neutrinos inferred from missing transverse momentum.\n",{"path":18475,"title":18476,"module":18468,"summary":18477},"\u002Fparticle-physics\u002Fexperiment\u002Fhow-discoveries-are-made","From Collisions to Discoveries","A discovery is a peak that survives statistics. Events are reconstructed into invariant masses, a signal accumulates as a bump over a smooth background, and its significance is judged by a p-value; the field's threshold is five sigma. The expected yield is a product — luminosity times cross section times branching ratio times acceptance and efficiency — that must be balanced by a trigger and data-reduction chain against an overwhelming rate. Worked reconstructions of $Z\\to\\ell\\ell$, the $J\u002F\\psi$, and the Higgs show the same peak-over-background logic at three scales.\n",{"path":18479,"title":18480,"module":18480,"summary":18481},"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fbeyond-standard-model","Beyond the Standard Model","The Standard Model leaves the four interactions ununified and the neutrinos massless, both now known to be wrong. Grand unification predicts the couplings merge near ten-to-the-sixteen GeV and the proton decays; supersymmetry pairs each particle with a superpartner; and the confirmed oscillation of neutrinos proves they carry mass, the first crack in the model.\n",{"path":18483,"title":18484,"module":18480,"summary":18485},"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fgrand-unified-theories","Grand Unified Theories and Proton Decay","The Standard Model gauge group is a product of three factors with three independent couplings. A grand unified theory embeds them in a single simple group — SU(5) is the minimal choice — so that one coupling runs into all three and the fractional quark charges follow from a tracelessness condition. The same embedding places quarks and leptons in shared multiplets, mediates baryon-number violation through superheavy gauge bosons, and predicts the proton decays with a lifetime that Super-Kamiokande has pushed past ten-to-the-thirty-four years.\n",{"path":18487,"title":18488,"module":18480,"summary":18489},"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fsupersymmetry","Supersymmetry","Supersymmetry relates fermions and bosons, pairing every Standard Model particle with a superpartner whose spin differs by one half. The pairing makes the scalar and fermion loop corrections to the Higgs mass cancel, removing the quadratic sensitivity to high scales; it sharpens the meeting of the three gauge couplings; and, when R-parity is conserved, it leaves the lightest superpartner stable and neutral, a natural dark-matter candidate. The LHC has excluded gluinos and light squarks below roughly two TeV.\n",{"path":18491,"title":18492,"module":18480,"summary":18493},"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fhierarchy-problem-naturalness","The Hierarchy Problem and Naturalness","The electroweak scale sits sixteen orders of magnitude below the Planck scale, and nothing in the Standard Model protects that gap. The Higgs mass squared picks up quadratic corrections proportional to the highest scale in the theory, so keeping it at the observed value requires the bare mass and its counterterm to cancel to some thirty significant figures. Naturalness treats that cancellation as a symptom of missing physics. Supersymmetry, compositeness, and extra dimensions each remove the quadratic sensitivity, but the LHC has found none of them at the predicted scale.\n",{"path":18495,"title":18496,"module":18480,"summary":18497},"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fdark-matter-candidates","Dark Matter and Particle Candidates","Flat galactic rotation curves, gravitational lensing, the cosmic microwave background, and structure formation all require about five times more matter than the visible baryons, none of it interacting electromagnetically. A stable weakly interacting particle of roughly weak-scale mass freezes out of the early universe with close to the observed abundance — the WIMP miracle — and is the leading candidate, with axions and sterile neutrinos as alternatives. Direct, indirect, and collider searches have so far only tightened the limits.\n",{"path":18499,"title":18500,"module":18480,"summary":18501},"\u002Fparticle-physics\u002Fbeyond-standard-model\u002Fmatter-antimatter-open-questions","Matter-Antimatter Asymmetry and Open Questions","The universe is made of matter, with about one extra baryon for every billion photons and no antimatter regions. Sakharov identified the three conditions any dynamical explanation must meet: baryon-number violation, C and CP violation, and a departure from thermal equilibrium. The Standard Model contains all three in principle, but its CP violation falls short by some ten orders of magnitude, so baryogenesis requires new physics — leptogenesis being the leading route. A closing survey collects the open questions and the experiments aimed at them.\n",{"path":18503,"title":18504,"module":6,"summary":6},"\u002Fparticle-physics","Particle Physics",{"path":18506,"title":18507,"module":18508,"summary":18509},"\u002Fastrophysics-cosmology\u002Forientation\u002Fthe-sun-and-stars","The Sun and the Life of Stars","Orientation","The Sun is the one star close enough to study in detail: its luminosity fixes a surface temperature of 5780 K, and the proton-proton fusion cycle in its 1.5-million-kelvin core supplies its power. Measuring other stars needs the magnitude scale, parallax, and the distance ladder; plotting luminosity against temperature builds the Hertzsprung-Russell diagram, on which a star's mass sets its lifetime and its evolutionary track off the main sequence.\n",{"path":18511,"title":18512,"module":18508,"summary":18513},"\u002Fastrophysics-cosmology\u002Forientation\u002Fstellar-death-final-states","Cataclysmic Events and the Final States of Stars","A star's death is set by its mass. In close binaries, matter poured across the Roche lobe onto a white dwarf produces novae and, at the Chandrasekhar limit of 1.4 solar masses, a Type Ia supernova; a massive star fusing to an iron core collapses into a Type II supernova. The remnant is a white dwarf held by electron degeneracy, a neutron star held by neutron degeneracy, or, above the neutron-star limit, a black hole inside its Schwarzschild radius.\n",{"path":18515,"title":18516,"module":18508,"summary":18517},"\u002Fastrophysics-cosmology\u002Forientation\u002Fgalaxies-and-cosmology","Galaxies, Cosmology, and the Evolving Universe","Galaxies come in elliptical, spiral, and irregular forms, and their redshifts obey Hubble's law, evidence that space itself is expanding. The critical density and the density parameter decide whether the universe is open, flat, or closed; baryons, dark matter, and dark energy each contribute. The cosmic microwave background and primordial helium anchor the Big Bang, whose thermal history runs from inflation through nucleosynthesis to the atoms of today.\n",{"path":18519,"title":18520,"module":18521,"summary":18522},"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Fmagnitudes-fluxes-and-the-distance-modulus","Magnitudes, Fluxes, and the Distance Modulus","Observational Foundations","The brightness of a star reaches us as a radiant flux that falls off as the inverse square of distance. The magnitude scale encodes flux logarithmically through the Pogson ratio; the apparent and absolute magnitudes differ by the distance modulus, which converts a measured brightness into a distance. The bolometric correction folds a filtered magnitude into a total luminosity, and the difference of two magnitudes in different bands, the color index, measures surface temperature.\n",{"path":18524,"title":18525,"module":18521,"summary":18526},"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Fstellar-spectra-and-spectral-classification","Stellar Spectra and Spectral Classification","A stellar spectrum is a continuum crossed by absorption lines whose strengths are set by the temperature of the atmosphere. The Boltzmann factor governs how atoms populate excited states, and the Saha equation governs how they ionize; their product explains why each line, such as the hydrogen Balmer series, peaks in strength at a characteristic temperature. This behavior orders stars into the OBAFGKM sequence, and the luminosity classes of the MK system add a second dimension for surface gravity.\n",{"path":18528,"title":18529,"module":18521,"summary":18530},"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Ftelescopes-and-detectors-across-the-spectrum","Telescopes and Detectors Across the Spectrum","A telescope collects light in proportion to its collecting area and resolves detail down to the diffraction limit set by its aperture and the observing wavelength. The atmosphere blurs and blocks large parts of the spectrum, which drives the choice between ground and space and between refractors, reflectors, and radio dishes. CCDs record the light with high quantum efficiency, and interferometry synthesizes an aperture as large as the separation of two telescopes.\n",{"path":18532,"title":18533,"module":18521,"summary":18534},"\u002Fastrophysics-cosmology\u002Fobservational-foundations\u002Fthe-cosmic-distance-ladder","The Cosmic Distance Ladder","No single method measures distances from the nearest stars to the far reaches of the universe. Instead a ladder of overlapping techniques, each calibrated by the one below it, extends the scale rung by rung: trigonometric parallax, main-sequence fitting, pulsating variables, the tip of the red-giant branch, the Tully-Fisher relation, and Type Ia supernovae. Each rung inherits the uncertainty of every rung beneath it, so the whole chain sets the accuracy of the Hubble constant.\n",{"path":18536,"title":18537,"module":18538,"summary":18539},"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fblackbody-radiation-and-specific-intensity","Blackbody Radiation and Specific Intensity","Radiation and Matter","Specific intensity is the fundamental measure of a radiation field: energy per unit area, time, frequency, and solid angle. It is conserved along a ray in empty space, and its angular moments give the mean intensity, flux, and radiation pressure. In thermal equilibrium the intensity equals the Planck function, whose limits and integrals reproduce the Rayleigh-Jeans law, the Wien law, Stefan-Boltzmann, and Wien's displacement law.\n",{"path":18541,"title":18542,"module":18538,"summary":18543},"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fradiative-transfer-and-the-transfer-equation","Radiative Transfer and the Transfer Equation","Along a ray, matter adds intensity through emission and removes it through absorption. Measuring path length in optical depth turns this into the transfer equation, whose formal solution superposes an attenuated background on the source function integrated along the line of sight. In local thermodynamic equilibrium the source function is the Planck function, and the Eddington-Barbier relation shows that the emergent intensity samples the source function at optical depth of order unity, explaining absorption lines and solar limb darkening.\n",{"path":18545,"title":18546,"module":18538,"summary":18547},"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fspectral-line-formation-and-broadening","Spectral-Line Formation and Broadening","A spectral line is a bound-bound transition whose strength is set by an oscillator strength and whose shape is set by three broadening mechanisms: the Lorentzian natural and collisional wings, the Gaussian thermal Doppler core, and their Voigt convolution. Equivalent width measures the total absorption, and the curve of growth relates it to the number of absorbers through a linear, saturated, and damping regime, turning line strengths into abundances.\n",{"path":18549,"title":18550,"module":18538,"summary":18551},"\u002Fastrophysics-cosmology\u002Fradiation-and-matter\u002Fopacity-and-the-rosseland-mean","Opacity Sources and the Rosseland Mean","Stellar opacity comes from four processes: bound-bound line absorption, bound-free photoionization, free-free absorption, and electron scattering. The bound-free and free-free terms follow a Kramers law, electron scattering sets a frequency-flat floor, and the negative hydrogen ion dominates cool photospheres. The Rosseland mean averages these harmonically, weighting transparent frequencies because they carry the flux, and its value fixes the radiative temperature gradient and decides where a star becomes convective.\n",{"path":18553,"title":18554,"module":18555,"summary":18556},"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fhydrostatic-equilibrium-and-the-virial-theorem","Hydrostatic Equilibrium and the Virial Theorem","Stellar Structure","A star holds itself up by balancing the inward pull of gravity against an outward pressure gradient. This balance, hydrostatic equilibrium, fixes a lower bound on the central pressure and, combined with the gravitational potential energy, yields the virial theorem. The virial relation gives a star a negative heat capacity, so that losing energy makes it hotter, and sets the Kelvin-Helmholtz timescale over which contraction alone can power the Sun.\n",{"path":18558,"title":18559,"module":18555,"summary":18560},"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fthe-equations-of-stellar-structure","The Equations of Stellar Structure","A static star is described by four coupled first-order differential equations in the interior mass or radius: mass conservation, hydrostatic equilibrium, energy generation, and energy transport. Closed with an equation of state, opacity, and reaction rates, and subject to central and surface boundary conditions, they determine the structure uniquely from mass and composition, the Vogt-Russell theorem. Energy moves by radiation until the temperature gradient exceeds the Schwarzschild limit, where convection takes over.\n",{"path":18562,"title":18563,"module":18555,"summary":18564},"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fthe-equation-of-state-and-polytropes","The Equation of State and Polytropes","Stellar pressure comes from gas, radiation, and, at high density, degenerate electrons. When pressure depends on density as a power law, hydrostatic equilibrium reduces to the Lane-Emden equation, whose solutions describe polytropes of index n. The relativistic degenerate case, n equal to three, gives a mass independent of radius, the Chandrasekhar mass. Eddington's standard model treats a radiation-supported star as an n equal to three polytrope and yields the quartic relating radiation fraction to mass.\n",{"path":18566,"title":18567,"module":18555,"summary":18568},"\u002Fastrophysics-cosmology\u002Fstellar-structure\u002Fthe-standard-solar-model","The Standard Solar Model","The standard solar model integrates the structure equations for one solar mass and calibrates the composition and convection parameter to reproduce the Sun's observed luminosity, radius, and age. Helioseismology tests the model's sound speed through the Sun's acoustic p-mode oscillations, and the model predicts a neutrino flux by production channel. The measured deficit, the solar-neutrino problem, is resolved by matter-enhanced flavor oscillation, confirmed when SNO measured the total flux across all flavors.\n",{"path":18570,"title":18571,"module":18572,"summary":18573},"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fthermonuclear-reaction-rates-and-the-gamow-peak","Thermonuclear Reaction Rates and the Gamow Peak","Nuclear Astrophysics","Stellar fusion proceeds only by quantum tunneling through the Coulomb barrier, because thermal energies are a thousand times smaller than the barrier height. The reaction rate is an integral over the Maxwell–Boltzmann distribution and the tunneling probability, whose product is sharply peaked at the Gamow energy. The astrophysical S-factor isolates the nuclear physics from the barrier penetration, and the steep temperature dependence follows from the width and position of the Gamow peak.\n",{"path":18575,"title":18576,"module":18572,"summary":18577},"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fhydrogen-burning-pp-chains-and-cno","Hydrogen Burning: pp Chains and the CNO Cycle","Four protons fuse into one helium-4 nucleus, releasing 26.7 MeV, through two competing networks. The pp chain begins with a weak-interaction bottleneck and branches three ways; the CNO cycle uses carbon, nitrogen, and oxygen as catalysts and is limited by nitrogen-14 proton capture. Their steep and gentle temperature dependences cross near 1.8e7 K, which divides pp-powered lower-main-sequence stars from CNO-powered upper-main-sequence stars.\n",{"path":18579,"title":18580,"module":18572,"summary":18581},"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fhelium-burning-and-the-triple-alpha-process","Helium Burning and the Triple-Alpha Process","Helium fuses to carbon in two steps through the unbound beryllium-8 nucleus and a resonant excited state of carbon-12, the Hoyle state, whose existence was predicted from the observed carbon abundance. The rate scales as roughly the fortieth power of temperature, and in a degenerate low-mass core this drives the runaway helium flash. A competing alpha capture on carbon-12 sets the carbon-to-oxygen ratio and the composition of the resulting white dwarf.\n",{"path":18583,"title":18584,"module":18572,"summary":18585},"\u002Fastrophysics-cosmology\u002Fnuclear-astrophysics\u002Fadvanced-burning-and-neutron-capture-nucleosynthesis","Advanced Burning, the Iron Peak, and the s\u002Fr Processes","Massive stars burn carbon, neon, oxygen, and silicon in ever-shorter stages, building an onion-shell interior and reaching nuclear statistical equilibrium at the iron peak, where the binding-energy-per-nucleon curve turns over and fusion can release no more energy. Elements beyond iron form by neutron capture: the slow s-process in AGB stars tracks the valley of stability, while the rapid r-process in supernovae and neutron-star mergers builds the heaviest nuclei far from it.\n",{"path":18587,"title":18588,"module":18589,"summary":18590},"\u002Fastrophysics-cosmology\u002Fism-and-star-formation\u002Fphases-of-the-interstellar-medium","The Phases of the Interstellar Medium","The Interstellar Medium","The gas between the stars separates into distinct thermal phases, from cold molecular clouds at 10 K to a diffuse million-degree corona, held near a common pressure by a balance of photoelectric heating and radiative cooling. Neutral hydrogen is traced by the 21-cm hyperfine line, dust reddens and extinguishes starlight along a characteristic wavelength law, and the ultraviolet output of hot stars carves ionized Strömgren spheres out of the surrounding gas.\n",{"path":18592,"title":18593,"module":18589,"summary":18594},"\u002Fastrophysics-cosmology\u002Fism-and-star-formation\u002Fmolecular-clouds-and-gravitational-collapse","Molecular Clouds and Gravitational Collapse","Stars form in cold, dense molecular clouds when self-gravity overcomes thermal and magnetic support. The virial theorem fixes the Jeans mass and length at which a clump becomes unstable, the free-fall time sets how fast it collapses, and a fragmentation cascade — cut off at a minimum mass by the onset of opacity — turns one cloud into a whole cluster, imprinting the stellar initial mass function.\n",{"path":18596,"title":18597,"module":18589,"summary":18598},"\u002Fastrophysics-cosmology\u002Fism-and-star-formation\u002Fprotostars-and-the-pre-main-sequence","Protostars and Pre-Main-Sequence Evolution","A collapsing core becomes optically thick and forms a protostar that grows by accretion through a disk while driving bipolar outflows. The newborn star appears on the birthline and contracts down the fully convective Hayashi track, then crosses the radiative Henyey track to the zero-age main sequence, powered by gravitational contraction until hydrogen ignites. Below about 0.08 solar masses degeneracy halts contraction before ignition, dividing stars from brown dwarfs.\n",{"path":18600,"title":18601,"module":18602,"summary":18603},"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fthe-main-sequence-and-its-structure","The Main Sequence and Its Structure","Stellar Evolution","A star settles onto the zero-age main sequence when core hydrogen ignition halts contraction. Homology scaling of the structure equations reproduces the mass–luminosity relation, and the burning mode splits the sequence into an upper branch with a convective core and a lower branch with a convective envelope. The main-sequence lifetime falls steeply with mass, and the turnoff of a coeval cluster serves as a clock.\n",{"path":18605,"title":18606,"module":18602,"summary":18607},"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fpost-main-sequence-low-mass-evolution","Post-Main-Sequence Evolution of Low-Mass Stars","When a low-mass star exhausts core hydrogen, burning moves to a shell, the core contracts, and the envelope swells into a red giant. A degenerate helium core ignites in a flash, settles onto the horizontal branch, and after a second contraction the star climbs the asymptotic giant branch with two burning shells. Thermal pulses and dredge-up enrich the surface, and mass loss ejects a planetary nebula, leaving a carbon–oxygen white dwarf.\n",{"path":18609,"title":18610,"module":18602,"summary":18611},"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fthe-evolution-of-massive-stars","The Evolution of Massive Stars","Stars above about eight solar masses burn through hydrogen, helium, carbon, neon, oxygen, and silicon in stages that grow shorter as neutrino losses accelerate contraction. The interior becomes an onion of concentric burning shells around an inert iron core. Radiation pressure near the Eddington limit drives fierce winds that can strip the hydrogen envelope entirely, and silicon burning builds an iron core toward the threshold of collapse.\n",{"path":18613,"title":18614,"module":18602,"summary":18615},"\u002Fastrophysics-cosmology\u002Fstellar-evolution\u002Fstellar-pulsation-and-the-instability-strip","Stellar Pulsation and the Instability Strip","Radial pulsation is a standing sound wave whose period scales inversely with the square root of the mean density. The kappa mechanism, an opacity valve seated in the helium partial-ionization zone, turns a star into a heat engine that pumps the oscillation. Stars in the instability strip pulsate as Cepheids, RR Lyrae, and Mira variables, and the Cepheid period–luminosity relation calibrates the distance ladder.\n",{"path":18617,"title":17039,"module":18618,"summary":18619},"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fwhite-dwarfs-and-the-chandrasekhar-limit","Stellar Death and Compact Remnants","A white dwarf is held up by the degeneracy pressure of its electrons, a quantum-mechanical stiffness that survives to zero temperature. Filling the Fermi sea sets a pressure that scales as density to the five-thirds power when the electrons are slow and only four-thirds when they are relativistic. The softer relativistic law produces the inverted mass-radius relation and a maximum mass, the Chandrasekhar limit near 1.4 solar masses, above which no cold equilibrium exists. Cooling and crystallization then turn the white-dwarf population into a clock for the Galactic disk.\n",{"path":18621,"title":18622,"module":18618,"summary":18623},"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fcore-collapse-supernovae","Core-Collapse Supernovae","When a massive star builds an iron core past the Chandrasekhar mass, degeneracy fails and the core collapses in less than a second. Photodisintegration and electron capture remove pressure support and neutronize the matter; the collapse halts abruptly at nuclear density, launching a shock that stalls and is revived by neutrino heating. The event is a Type II or stripped-envelope Ib\u002FIc supernova, and the neutrinos from SN 1987A confirmed the picture directly.\n",{"path":18625,"title":18626,"module":18618,"summary":18627},"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fthermonuclear-supernovae-type-ia","Thermonuclear Supernovae","A carbon-oxygen white dwarf driven toward the Chandrasekhar mass ignites its degenerate fuel and unbinds itself in a thermonuclear runaway, the Type Ia supernova. The light curve is powered by the radioactive decay of nickel-56 to cobalt-56 to iron-56, and the Phillips relation between peak brightness and decline rate makes these events standardizable candles. Their near-uniform luminosity turns them into the distance indicators that revealed cosmic acceleration.\n",{"path":18629,"title":18630,"module":18618,"summary":18631},"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fneutron-stars-and-pulsars","Neutron Stars and Pulsars","A neutron star is held up by neutron degeneracy and the repulsive nuclear force, with a maximum mass, the Tolman-Oppenheimer-Volkoff limit, set by an uncertain dense-matter equation of state. Its rotating magnetic dipole sweeps a beam past Earth as a pulsar, and magnetic braking traces a track across the period-period- derivative diagram. Millisecond pulsars, magnetars, glitches, and the orbital decay of the Hulse-Taylor binary follow from the same structure.\n",{"path":18633,"title":18634,"module":18618,"summary":18635},"\u002Fastrophysics-cosmology\u002Fstellar-death-and-compact-remnants\u002Fblack-holes-schwarzschild-and-kerr","Black Holes, Schwarzschild and Kerr","Above the neutron-star mass limit gravity wins completely and the remnant is a black hole. The Schwarzschild solution gives the event horizon, gravitational redshift, and time dilation; the innermost stable circular orbit sets the efficiency of accretion. Rotating Kerr black holes drag spacetime and carry an ergosphere. Stellar-mass black holes are found in X-ray binaries, and the Event Horizon Telescope has imaged the shadow of a supermassive one.\n",{"path":18637,"title":18638,"module":18639,"summary":18640},"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Fbinary-systems-and-mass-transfer","Binary Systems and Mass Transfer","Binaries and Gravitational Waves","Most stars are born in pairs, and a binary is the only setting where a stellar mass can be measured directly. Visual, spectroscopic, and eclipsing binaries each expose a different combination of the orbital elements, and together they calibrate the mass-luminosity relation. When one star swells to fill its Roche lobe, gas streams through the inner Lagrange point onto its companion. Conservative transfer widens or shrinks the orbit depending on the mass ratio, and the sign of that response explains the Algol paradox.\n",{"path":18642,"title":18643,"module":18639,"summary":18644},"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Faccreting-compact-objects","Accreting Compact Objects","Gas falling onto a compact object converts gravitational binding energy into radiation with an efficiency set by the depth of the potential well, up to tens of percent of the rest mass for a neutron star or black hole. Angular momentum forces the flow into a disk, and viscous dissipation gives a temperature profile that falls as radius to the minus three-quarters, producing a multicolor blackbody spectrum. Radiation pressure caps the steady luminosity at the Eddington limit. Unstable nuclear burning of the accreted fuel powers classical novae on white dwarfs and Type I X-ray bursts on neutron stars.\n",{"path":18646,"title":18647,"module":18639,"summary":18648},"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Fgravitational-waves-from-inspiraling-binaries","Gravitational Waves from Inspiraling Binaries","A time-varying mass quadrupole radiates gravitational waves, ripples in spacetime that stretch and squeeze a ring of free masses along two polarizations. The radiated power drains a binary's orbital energy, shrinking the orbit and sweeping the wave frequency upward in a chirp whose rate fixes the chirp mass. Laser interferometers with kilometre arms measure the resulting strain of order ten to the minus twenty-one. The first detection, GW150914, matched a template for two merging black holes near thirty solar masses each.\n",{"path":18650,"title":18651,"module":18639,"summary":18652},"\u002Fastrophysics-cosmology\u002Fbinaries-and-gravitational-waves\u002Fmultimessenger-astronomy-and-gamma-ray-bursts","Multimessenger Astronomy and Gamma-Ray Bursts","Gamma-ray bursts split into two populations: long bursts from the collapse of massive stars and short bursts from merging compact objects. The compactness problem forces the emitting plasma to move at ultra-relativistic speed, beaming the radiation into a narrow jet. The neutron-star merger GW170817 tied a gravitational chirp to a short gamma-ray burst, a radioactive kilonova, and a broadband afterglow, confirming that mergers forge r-process elements. A merger with a measured redshift is a standard siren that reads the Hubble constant from gravitational data alone.\n",{"path":18654,"title":18655,"module":18656,"summary":18657},"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fthe-milky-way","The Milky Way Galaxy","Galaxies and Dark Matter","The Galaxy resolves into a thin disk of gas and young stars, a central bar and bulge, and a diffuse old halo studded with globular clusters. Star counts and the reddening of distant light map these components, while the differential rotation of the disk — encoded in the Oort constants and the flat rotation curve — measures the enclosed mass and reveals more than the stars can account for. Spiral arms are density waves, not material structures, and the innermost stellar orbits around Sgr A* weigh a four-million-solar-mass black hole.\n",{"path":18659,"title":18660,"module":18656,"summary":18661},"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fgalaxy-morphology-and-classification","Galaxy Morphology and Classification","Galaxies sort along the Hubble tuning fork from smooth ellipticals through lenticulars to grand-design and barred spirals, with irregulars off the end. The light of a spheroid follows the de Vaucouleurs quarter-power law while a disk fades exponentially, and the general Sérsic profile interpolates between them. Virial scaling relations — Tully–Fisher for disks, Faber–Jackson and the fundamental plane for spheroids — tie luminosity to internal motions, and the Schechter function fixes the abundance of galaxies as a function of luminosity.\n",{"path":18663,"title":18664,"module":18656,"summary":18665},"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fgalaxy-rotation-curves-and-dark-matter","Galaxy Rotation Curves and Dark Matter","The rotation curves of disk galaxies stay flat far beyond the light, demanding an extended halo whose density falls as the inverse square of radius. Decomposing the curve into disk, bulge, and halo, and fitting isothermal or NFW profiles, quantifies the missing mass. Gravitational lensing weighs the same mass without dynamics, the mass-to-light ratio climbs from stars to clusters, and the Bullet Cluster separates the collisionless dark matter from the colliding gas — evidence that MOND strains to match.\n",{"path":18667,"title":18668,"module":18656,"summary":18669},"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Factive-galactic-nuclei-and-supermassive-black-holes","Active Galactic Nuclei","A small fraction of galaxies pour out enormous luminosity from a region smaller than the solar system. Accretion onto a supermassive black hole, limited by the Eddington balance of radiation pressure and gravity, powers the Seyferts, quasars, radio galaxies, and blazars — one engine seen from different angles through an obscuring torus. Relativistic jets produce apparent superluminal motion, reverberation mapping and stellar dynamics weigh the central mass, and the M–sigma relation ties that mass to the host bulge.\n",{"path":18671,"title":18672,"module":18656,"summary":18673},"\u002Fastrophysics-cosmology\u002Fgalaxies\u002Fgalaxy-clusters-and-large-scale-structure","Galaxy Clusters and Large-Scale Structure","Galaxies gather into groups and rich clusters bound by a common dark halo and filled with hot X-ray gas. Three independent probes — the virial theorem, the hydrostatic X-ray temperature, and gravitational lensing — agree on a mass that dwarfs the stars. On the largest scales galaxies trace a cosmic web of filaments, walls, and voids, quantified by the two-point correlation function, whose baryon acoustic oscillation bump provides a standard ruler for cosmology.\n",{"path":18675,"title":18676,"module":18677,"summary":18678},"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fthe-expanding-universe-and-hubbles-law","The Expanding Universe and Hubble's Law","Cosmic Expansion and Dynamics","The universe is homogeneous and isotropic on large scales, so its expansion is captured by a single function of time, the scale factor. Comoving coordinates stay fixed while proper distances grow in proportion to the scale factor, producing Hubble's law and a cosmological redshift that measures stretched space rather than a Doppler shift. A Newtonian energy argument reproduces the dynamics, and the same finite, expanding cosmos resolves Olbers' paradox.\n",{"path":18680,"title":18681,"module":18677,"summary":18682},"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fthe-frw-metric-and-cosmological-redshift","The FRW Metric and Cosmological Redshift","The geometry of a homogeneous, isotropic universe is fixed by symmetry to the Robertson-Walker metric, with the entire freedom reduced to a scale factor and a single curvature constant selecting an open, flat, or closed space. From the metric the null geodesic of light gives comoving distance, the exact cosmological redshift, and the distinction between the proper distance we cannot measure and the redshift we can.\n",{"path":18684,"title":15760,"module":18677,"summary":18685},"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fthe-friedmann-equations-and-cosmic-dynamics","The scale factor obeys the Friedmann equation, the acceleration equation, and the fluid equation, only two of which are independent. An equation of state fixes how each component behaves under expansion, so radiation dilutes as the inverse fourth power of the scale factor, matter as the inverse cube, and vacuum energy not at all. The critical density defines the density parameters, and the deceleration parameter encodes whether gravity or dark energy is winning.\n",{"path":18687,"title":18688,"module":18677,"summary":18689},"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fcosmological-models-and-distances","Cosmological Models and Distances","Integrating the Friedmann equation for particular mixtures gives the benchmark models, from the matter-only Einstein-de Sitter universe to the concordance Lambda-CDM, each with its own scale-factor history and age. Because the redshift is the only direct observable, several distance measures diverge at high redshift, and the angular-diameter distance even turns over so that the most distant objects look larger. The horizon and lookback time set what is causally and observationally reachable.\n",{"path":18691,"title":18692,"module":18677,"summary":18693},"\u002Fastrophysics-cosmology\u002Fcosmology-expansion-and-dynamics\u002Fdark-energy-and-the-accelerating-universe","Dark Energy and the Accelerating Universe","In 1998 two teams found that distant Type Ia supernovae are fainter than a decelerating universe predicts, revealing that the expansion is accelerating and that a component with negative pressure dominates the energy budget. The simplest candidate is the cosmological constant, or vacuum energy, with an equation of state near minus one. It works observationally but leaves two deep puzzles: why the vacuum energy is a hundred and twenty orders of magnitude smaller than expected, and why it is comparable to the matter density just now.\n",{"path":18695,"title":18696,"module":18697,"summary":18698},"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fthe-thermal-history-of-the-universe","The Thermal History of the Universe","The Hot Big Bang","Running the expansion backward compresses and heats the universe, so its past is a sequence of thermal epochs set by temperature. Temperature scales as the inverse scale factor; species stay in equilibrium while their interaction rate exceeds the expansion rate and freeze out when it drops below. The effective degrees of freedom count the relativistic species and step down through mass thresholds, and neutrino decoupling just before electron-positron annihilation leaves a relic neutrino background slightly cooler than the photons.\n",{"path":18700,"title":18701,"module":18697,"summary":18702},"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fbig-bang-nucleosynthesis","Big Bang Nucleosynthesis","In the first three minutes the weak interactions freeze out the neutron-to-proton ratio, and once deuterium survives photodissociation a fast reaction network converts nearly all free neutrons into helium-4. The primordial abundances of deuterium, helium-3, helium-4, and lithium-7 depend on a single free parameter, the baryon-to-photon ratio, so measuring them fixes the baryon density. The predictions match observation across nine decades of abundance, with a persistent discrepancy in lithium-7.\n",{"path":18704,"title":18705,"module":18697,"summary":18706},"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Frecombination-and-the-cosmic-microwave-background","Recombination and the Cosmic Microwave Background","As the universe cooled through a few thousand kelvin the free electrons bound to protons, and the Saha equation tracks the falling ionization fraction. Once the plasma neutralized, photons stopped scattering and streamed freely from a spherical surface of last scattering at redshift about 1100. Those photons are the cosmic microwave background, an almost perfect blackbody at 2.725 kelvin with a dipole from our motion through it.\n",{"path":18708,"title":18709,"module":18697,"summary":18710},"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fcmb-anisotropies-and-cosmological-parameters","CMB Anisotropies and Cosmological Parameters","The cosmic microwave background carries temperature fluctuations at the ten-parts-per-million level, imprinted by sound waves in the photon-baryon plasma before recombination. Decomposed into spherical harmonics, the fluctuations form an angular power spectrum whose acoustic peaks encode the geometry and contents of the universe: the first peak fixes spatial flatness, the odd-even peak ratio the baryon density, and the third peak the dark-matter density. Polarization adds an independent channel, and the Planck measurements pin the concordance parameters.\n",{"path":18712,"title":18713,"module":18697,"summary":18714},"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fcosmic-inflation","Cosmic Inflation","The hot Big Bang leaves three initial-condition puzzles unexplained: why causally disconnected patches share a temperature, why the geometry is so nearly flat, and why no magnetic monopoles are seen. A brief epoch of accelerated expansion driven by a slowly rolling scalar field solves all three by stretching a small causal patch across the observable universe. The same accelerated expansion freezes quantum fluctuations into a near-scale-invariant spectrum of density perturbations, seeding all later structure.\n",{"path":18716,"title":18717,"module":18697,"summary":18718},"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fstructure-formation-and-the-growth-of-perturbations","Structure Formation and the Growth of Perturbations","The near-uniform early universe grew its galaxies and clusters by gravitational instability acting on the tiny inflationary perturbations. In an expanding background the growth is slowed to a power law rather than the exponential of a static medium; perturbations stall during radiation domination and grow with the scale factor once matter dominates. The transfer function turns the primordial spectrum into the processed matter power spectrum, and cold dark matter builds structure from the bottom up.\n",{"path":18720,"title":18721,"module":18697,"summary":18722},"\u002Fastrophysics-cosmology\u002Fthe-hot-big-bang\u002Fdark-matter-dark-energy-and-open-questions","Dark Matter, Dark Energy, and Open Questions","Five independent lines of evidence converge on a universe whose energy budget is dominated by dark energy and dark matter, with ordinary baryons a small remainder. The candidate particles for dark matter range from WIMPs to axions to sterile neutrinos, each with its own detection strategy. The concordance model fits the data with six parameters but leaves the nature of dark energy, the Hubble tension, small-scale structure, and the matter-antimatter asymmetry unexplained.\n",{"path":18724,"title":18725,"module":6,"summary":6},"\u002Fastrophysics-cosmology","Astrophysics & Cosmology",{"path":18727,"title":18728,"module":6,"summary":6},"\u002Fcolophon","Colophon",{"path":7876,"title":18730,"module":6,"summary":6},"Study Notes",[18732,18745,18757,18766,18775,18790,18847,18866],{"module":14054,"moduleNumber":2097,"slug":18733,"lessons":18734},"foundations",[18735,18737,18739,18741,18743],{"title":18095,"path":18094,"lessonNumber":2097,"topics":18736,"summary":18096},[14054],{"title":18099,"path":18098,"lessonNumber":2103,"topics":18738,"summary":18100},[14054],{"title":18103,"path":18102,"lessonNumber":2109,"topics":18740,"summary":18104},[14054],{"title":18107,"path":18106,"lessonNumber":2115,"topics":18742,"summary":18108},[14054],{"title":18111,"path":18110,"lessonNumber":2121,"topics":18744,"summary":18112},[14054],{"module":18116,"moduleNumber":2103,"slug":18746,"lessons":18747},"classification",[18748,18751,18753,18755],{"title":18115,"path":18114,"lessonNumber":2097,"topics":18749,"summary":18117},[18750],"Classification",{"title":18120,"path":18119,"lessonNumber":2103,"topics":18752,"summary":18121},[18750],{"title":18124,"path":18123,"lessonNumber":2109,"topics":18754,"summary":18125},[18750],{"title":18128,"path":18127,"lessonNumber":2115,"topics":18756,"summary":18129},[18750],{"module":18133,"moduleNumber":2109,"slug":18758,"lessons":18759},"semantics",[18760,18762,18764],{"title":18132,"path":18131,"lessonNumber":2097,"topics":18761,"summary":18134},[18133],{"title":18137,"path":18136,"lessonNumber":2103,"topics":18763,"summary":18138},[18133],{"title":18140,"path":17,"lessonNumber":2109,"topics":18765,"summary":18141},[18133],{"module":13116,"moduleNumber":2115,"slug":18767,"lessons":18768},"sequences",[18769,18771,18773],{"title":18143,"path":5656,"lessonNumber":2097,"topics":18770,"summary":18144},[13116],{"title":18147,"path":18146,"lessonNumber":2103,"topics":18772,"summary":18148},[13116],{"title":5,"path":13118,"lessonNumber":2109,"topics":18774,"summary":13135},[13116],{"module":15001,"moduleNumber":2121,"slug":18776,"lessons":18777},"transformers",[18778,18780,18782,18784,18786,18788],{"title":18151,"path":11834,"lessonNumber":2097,"topics":18779,"summary":18152},[15001],{"title":16710,"path":18154,"lessonNumber":2103,"topics":18781,"summary":18155},[15001],{"title":16818,"path":18157,"lessonNumber":2109,"topics":18783,"summary":18158},[15001],{"title":18161,"path":18160,"lessonNumber":2115,"topics":18785,"summary":18162},[15001],{"title":18165,"path":18164,"lessonNumber":2121,"topics":18787,"summary":18166},[15001],{"title":18169,"path":18168,"lessonNumber":2127,"topics":18789,"summary":18170},[15001],{"module":18174,"moduleNumber":2127,"slug":18791,"lessons":18792},"linguistic-structure",[18793,18796,18798,18800,18802,18804,18806,18808,18811,18814,18817,18820,18823,18826,18829,18832,18835,18838,18841,18844],{"title":18173,"path":18172,"lessonNumber":2097,"topics":18794,"summary":18175},[18795],"Structure",{"title":18178,"path":18177,"lessonNumber":2103,"topics":18797,"summary":18179},[18795],{"title":18182,"path":18181,"lessonNumber":2109,"topics":18799,"summary":18183},[18795],{"title":18186,"path":18185,"lessonNumber":2115,"topics":18801,"summary":18187},[18795],{"title":18190,"path":18189,"lessonNumber":2121,"topics":18803,"summary":18191},[18795],{"title":18194,"path":18193,"lessonNumber":2127,"topics":18805,"summary":18195},[18795],{"title":18198,"path":18197,"lessonNumber":2133,"topics":18807,"summary":18199},[18795],{"title":18202,"path":18201,"lessonNumber":18809,"topics":18810,"summary":18203},8,[18795],{"title":18206,"path":18205,"lessonNumber":18812,"topics":18813,"summary":18207},9,[18795],{"title":18210,"path":18209,"lessonNumber":18815,"topics":18816,"summary":18211},10,[18795],{"title":18214,"path":18213,"lessonNumber":18818,"topics":18819,"summary":18215},11,[18795],{"title":18218,"path":18217,"lessonNumber":18821,"topics":18822,"summary":18219},12,[18795],{"title":18222,"path":18221,"lessonNumber":18824,"topics":18825,"summary":18223},13,[18795],{"title":18226,"path":18225,"lessonNumber":18827,"topics":18828,"summary":18227},14,[18795],{"title":18230,"path":18229,"lessonNumber":18830,"topics":18831,"summary":18231},15,[18795],{"title":18234,"path":18233,"lessonNumber":18833,"topics":18834,"summary":18235},16,[18795],{"title":18238,"path":18237,"lessonNumber":18836,"topics":18837,"summary":18239},17,[18795],{"title":18242,"path":18241,"lessonNumber":18839,"topics":18840,"summary":18243},18,[18795],{"title":18246,"path":18245,"lessonNumber":18842,"topics":18843,"summary":18247},19,[18795],{"title":18250,"path":18249,"lessonNumber":18845,"topics":18846,"summary":18251},20,[18795],{"module":16806,"moduleNumber":2133,"slug":18848,"lessons":18849},"applications",[18850,18852,18854,18856,18858,18860,18862,18864],{"title":18254,"path":18253,"lessonNumber":2097,"topics":18851,"summary":18255},[16806],{"title":18258,"path":18257,"lessonNumber":2103,"topics":18853,"summary":18259},[16806],{"title":18262,"path":18261,"lessonNumber":2109,"topics":18855,"summary":18263},[16806],{"title":18266,"path":18265,"lessonNumber":2115,"topics":18857,"summary":18267},[16806],{"title":18270,"path":18269,"lessonNumber":2121,"topics":18859,"summary":18271},[16806],{"title":18274,"path":18273,"lessonNumber":2127,"topics":18861,"summary":18275},[16806],{"title":18278,"path":18277,"lessonNumber":2133,"topics":18863,"summary":18279},[16806],{"title":18282,"path":18281,"lessonNumber":18809,"topics":18865,"summary":18283},[16806],{"module":18287,"moduleNumber":18809,"slug":18867,"lessons":18868},"speech",[18869,18871,18873,18875],{"title":18286,"path":18285,"lessonNumber":2097,"topics":18870,"summary":18288},[18287],{"title":18291,"path":18290,"lessonNumber":2103,"topics":18872,"summary":18292},[18287],{"title":18295,"path":18294,"lessonNumber":2109,"topics":18874,"summary":18296},[18287],{"title":18299,"path":18298,"lessonNumber":2115,"topics":18876,"summary":18300},[18287],"\u003Csvg style=\"width:100%;max-width:193.651px;height:auto\" xmlns=\"http:\u002F\u002Fwww.w3.org\u002F2000\u002Fsvg\" viewBox=\"-75 -75 145.239 140.128\">\u003Cg stroke=\"currentColor\" style=\"stroke-miterlimit:10;stroke-width:.4\">\u003Cpath fill=\"none\" d=\"M29.58 61.658h36.989V36.05H29.58Z\"\u002F>\u003Cg transform=\"translate(-7.837 2.25)\">\u003Cpath d=\"M49.876 48.854L48.201 48.854L48.201 48.538Q48.544 48.538 48.850 48.397Q49.155 48.256 49.370 47.988L50.197 46.973L49.129 45.589Q48.957 45.387 48.775 45.338Q48.593 45.290 48.245 45.290L48.245 44.974L50.144 44.974L50.144 45.290Q50.034 45.290 49.939 45.345Q49.845 45.400 49.845 45.497Q49.845 45.536 49.876 45.589L50.583 46.494L51.124 45.839Q51.234 45.699 51.234 45.549Q51.234 45.439 51.166 45.365Q51.097 45.290 50.992 45.290L50.992 44.974L52.662 44.974L52.662 45.290Q52.319 45.290 52.011 45.433Q51.704 45.576 51.493 45.839L50.763 46.731L51.946 48.243Q52.064 48.384 52.194 48.448Q52.324 48.511 52.455 48.524Q52.587 48.538 52.824 48.538L52.824 48.854L50.917 48.854L50.917 48.538Q51.023 48.538 51.124 48.483Q51.225 48.428 51.225 48.331Q51.225 48.283 51.199 48.243L50.386 47.197L49.740 47.988Q49.630 48.138 49.630 48.278Q49.630 48.384 49.702 48.461Q49.775 48.538 49.876 48.538L49.876 48.854M55.804 51.095Q55.299 50.708 54.929 50.203Q54.560 49.698 54.316 49.098Q54.073 48.498 53.958 47.881Q53.844 47.263 53.844 46.604Q53.844 45.945 53.958 45.330Q54.073 44.714 54.312 44.121Q54.552 43.528 54.925 43.018Q55.299 42.508 55.804 42.122Q55.839 42.104 55.861 42.104L55.940 42.104Q56.028 42.104 56.028 42.205Q56.028 42.240 55.993 42.275Q55.430 42.798 55.085 43.504Q54.741 44.209 54.593 44.991Q54.446 45.773 54.446 46.604Q54.446 47.228 54.525 47.812Q54.604 48.397 54.782 48.964Q54.960 49.531 55.259 50.032Q55.558 50.533 55.993 50.941Q56.028 50.977 56.028 51.016Q56.028 51.113 55.940 51.113L55.861 51.113Q55.839 51.113 55.804 51.095M57.478 47.782L57.478 45.290L56.714 45.290L56.714 45.031Q57.118 45.031 57.384 44.765Q57.650 44.499 57.771 44.099Q57.891 43.699 57.891 43.317L58.181 43.317L58.181 44.974L59.469 44.974L59.469 45.290L58.181 45.290L58.181 47.747Q58.181 48.116 58.307 48.390Q58.432 48.665 58.757 48.665Q59.056 48.665 59.194 48.371Q59.333 48.076 59.333 47.747L59.333 47.224L59.618 47.224L59.618 47.782Q59.618 48.059 59.509 48.331Q59.399 48.604 59.186 48.779Q58.972 48.955 58.691 48.955Q58.331 48.955 58.058 48.817Q57.786 48.678 57.632 48.415Q57.478 48.151 57.478 47.782M60.818 51.113L60.735 51.113Q60.647 51.113 60.647 51.016Q60.647 50.977 60.682 50.941Q61.517 50.168 61.877 49.034Q62.238 47.900 62.238 46.604Q62.238 45.984 62.158 45.393Q62.079 44.802 61.899 44.244Q61.719 43.686 61.418 43.178Q61.117 42.671 60.682 42.275Q60.647 42.240 60.647 42.205Q60.647 42.104 60.735 42.104L60.818 42.104Q60.836 42.104 60.871 42.122Q61.376 42.504 61.750 43.018Q62.123 43.532 62.361 44.110Q62.598 44.688 62.714 45.316Q62.831 45.945 62.831 46.604Q62.831 47.263 62.714 47.894Q62.598 48.524 62.358 49.109Q62.119 49.693 61.748 50.203Q61.376 50.713 60.871 51.095Q60.836 51.113 60.818 51.113\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M29.58 7.598h36.989V-18.01H29.58Z\"\u002F>\u003Cg transform=\"translate(-7.965 -51.81)\">\u003Cpath d=\"M50.460 48.854L48.373 48.854L48.373 48.538Q48.680 48.538 48.872 48.485Q49.063 48.432 49.063 48.243L49.063 43.528Q49.063 43.286 48.992 43.178Q48.922 43.071 48.788 43.047Q48.654 43.022 48.373 43.022L48.373 42.706L49.740 42.609L49.740 45.659Q49.937 45.303 50.291 45.090Q50.645 44.877 51.045 44.877Q52.324 44.877 52.324 46.090L52.324 48.243Q52.324 48.432 52.515 48.485Q52.706 48.538 53.013 48.538L53.013 48.854L50.926 48.854L50.926 48.538Q51.238 48.538 51.429 48.485Q51.620 48.432 51.620 48.243L51.620 46.125Q51.620 45.866 51.576 45.644Q51.532 45.422 51.387 45.279Q51.242 45.136 50.983 45.136Q50.640 45.136 50.359 45.325Q50.078 45.514 49.922 45.826Q49.766 46.138 49.766 46.485L49.766 48.243Q49.766 48.432 49.959 48.485Q50.153 48.538 50.460 48.538L50.460 48.854M56.063 51.095Q55.558 50.708 55.189 50.203Q54.820 49.698 54.576 49.098Q54.332 48.498 54.218 47.881Q54.103 47.263 54.103 46.604Q54.103 45.945 54.218 45.330Q54.332 44.714 54.571 44.121Q54.811 43.528 55.184 43.018Q55.558 42.508 56.063 42.122Q56.098 42.104 56.120 42.104L56.199 42.104Q56.287 42.104 56.287 42.205Q56.287 42.240 56.252 42.275Q55.690 42.798 55.345 43.504Q55 44.209 54.853 44.991Q54.705 45.773 54.705 46.604Q54.705 47.228 54.784 47.812Q54.864 48.397 55.042 48.964Q55.220 49.531 55.518 50.032Q55.817 50.533 56.252 50.941Q56.287 50.977 56.287 51.016Q56.287 51.113 56.199 51.113L56.120 51.113Q56.098 51.113 56.063 51.095M57.738 47.782L57.738 45.290L56.973 45.290L56.973 45.031Q57.377 45.031 57.643 44.765Q57.909 44.499 58.030 44.099Q58.151 43.699 58.151 43.317L58.441 43.317L58.441 44.974L59.728 44.974L59.728 45.290L58.441 45.290L58.441 47.747Q58.441 48.116 58.566 48.390Q58.691 48.665 59.016 48.665Q59.315 48.665 59.454 48.371Q59.592 48.076 59.592 47.747L59.592 47.224L59.878 47.224L59.878 47.782Q59.878 48.059 59.768 48.331Q59.658 48.604 59.445 48.779Q59.232 48.955 58.950 48.955Q58.590 48.955 58.318 48.817Q58.045 48.678 57.891 48.415Q57.738 48.151 57.738 47.782M61.077 51.113L60.994 51.113Q60.906 51.113 60.906 51.016Q60.906 50.977 60.941 50.941Q61.776 50.168 62.136 49.034Q62.497 47.900 62.497 46.604Q62.497 45.984 62.418 45.393Q62.339 44.802 62.158 44.244Q61.978 43.686 61.677 43.178Q61.376 42.671 60.941 42.275Q60.906 42.240 60.906 42.205Q60.906 42.104 60.994 42.104L61.077 42.104Q61.095 42.104 61.130 42.122Q61.636 42.504 62.009 43.018Q62.383 43.532 62.620 44.110Q62.857 44.688 62.974 45.316Q63.090 45.945 63.090 46.604Q63.090 47.263 62.974 47.894Q62.857 48.524 62.618 49.109Q62.378 49.693 62.007 50.203Q61.636 50.713 61.130 51.095Q61.095 51.113 61.077 51.113\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M29.58-46.463h36.989V-72.07H29.58Z\"\u002F>\u003Cg transform=\"translate(-7.837 -105.87)\">\u003Cpath d=\"M48.751 50.331Q48.918 50.436 49.098 50.436Q49.423 50.436 49.660 50.192Q49.898 49.948 50.047 49.601L50.350 48.854L48.984 45.589Q48.891 45.391 48.729 45.341Q48.566 45.290 48.263 45.290L48.263 44.974L50.188 44.974L50.188 45.290Q49.696 45.290 49.696 45.505Q49.696 45.527 49.713 45.589L50.720 47.979L51.620 45.839Q51.656 45.756 51.656 45.650Q51.656 45.488 51.535 45.389Q51.414 45.290 51.251 45.290L51.251 44.974L52.754 44.974L52.754 45.290Q52.469 45.290 52.255 45.433Q52.042 45.576 51.937 45.839L50.350 49.601Q50.161 50.054 49.847 50.377Q49.533 50.700 49.098 50.700Q48.773 50.700 48.513 50.480Q48.254 50.260 48.254 49.935Q48.254 49.768 48.373 49.654Q48.491 49.540 48.658 49.540Q48.773 49.540 48.863 49.590Q48.953 49.641 49.003 49.729Q49.054 49.816 49.054 49.935Q49.054 50.080 48.973 50.190Q48.891 50.300 48.751 50.331M55.804 51.095Q55.299 50.708 54.929 50.203Q54.560 49.698 54.316 49.098Q54.073 48.498 53.958 47.881Q53.844 47.263 53.844 46.604Q53.844 45.945 53.958 45.330Q54.073 44.714 54.312 44.121Q54.552 43.528 54.925 43.018Q55.299 42.508 55.804 42.122Q55.839 42.104 55.861 42.104L55.940 42.104Q56.028 42.104 56.028 42.205Q56.028 42.240 55.993 42.275Q55.430 42.798 55.085 43.504Q54.741 44.209 54.593 44.991Q54.446 45.773 54.446 46.604Q54.446 47.228 54.525 47.812Q54.604 48.397 54.782 48.964Q54.960 49.531 55.259 50.032Q55.558 50.533 55.993 50.941Q56.028 50.977 56.028 51.016Q56.028 51.113 55.940 51.113L55.861 51.113Q55.839 51.113 55.804 51.095M57.478 47.782L57.478 45.290L56.714 45.290L56.714 45.031Q57.118 45.031 57.384 44.765Q57.650 44.499 57.771 44.099Q57.891 43.699 57.891 43.317L58.181 43.317L58.181 44.974L59.469 44.974L59.469 45.290L58.181 45.290L58.181 47.747Q58.181 48.116 58.307 48.390Q58.432 48.665 58.757 48.665Q59.056 48.665 59.194 48.371Q59.333 48.076 59.333 47.747L59.333 47.224L59.618 47.224L59.618 47.782Q59.618 48.059 59.509 48.331Q59.399 48.604 59.186 48.779Q58.972 48.955 58.691 48.955Q58.331 48.955 58.058 48.817Q57.786 48.678 57.632 48.415Q57.478 48.151 57.478 47.782M60.818 51.113L60.735 51.113Q60.647 51.113 60.647 51.016Q60.647 50.977 60.682 50.941Q61.517 50.168 61.877 49.034Q62.238 47.900 62.238 46.604Q62.238 45.984 62.158 45.393Q62.079 44.802 61.899 44.244Q61.719 43.686 61.418 43.178Q61.117 42.671 60.682 42.275Q60.647 42.240 60.647 42.205Q60.647 42.104 60.735 42.104L60.818 42.104Q60.836 42.104 60.871 42.122Q61.376 42.504 61.750 43.018Q62.123 43.532 62.361 44.110Q62.598 44.688 62.714 45.316Q62.831 45.945 62.831 46.604Q62.831 47.263 62.714 47.894Q62.598 48.524 62.358 49.109Q62.119 49.693 61.748 50.203Q61.376 50.713 60.871 51.095Q60.836 51.113 60.818 51.113\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M48.074 35.85V10.598\"\u002F>\u003Cpath stroke=\"none\" d=\"m48.074 7.998-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003Cg transform=\"translate(3.733 -24.538)\">\u003Cpath d=\"M50.467 48.895L48.915 44.564Q48.853 44.421 48.691 44.387Q48.529 44.353 48.276 44.353L48.276 44.072L50.203 44.072L50.203 44.353Q49.622 44.353 49.622 44.527Q49.622 44.547 49.629 44.564L50.853 48L51.960 44.913L51.834 44.564Q51.776 44.421 51.610 44.387Q51.444 44.353 51.195 44.353L51.195 44.072L53.122 44.072L53.122 44.353Q52.541 44.353 52.541 44.527L52.541 44.564L53.772 48L54.927 44.752Q54.941 44.711 54.941 44.687Q54.941 44.513 54.748 44.433Q54.554 44.353 54.346 44.353L54.346 44.072L55.922 44.072L55.922 44.353Q55.672 44.353 55.481 44.447Q55.289 44.541 55.214 44.752L53.731 48.895Q53.697 48.994 53.597 48.994L53.519 48.994Q53.420 48.994 53.379 48.895L52.100 45.306L50.819 48.895Q50.802 48.939 50.764 48.967Q50.726 48.994 50.678 48.994L50.600 48.994Q50.508 48.994 50.467 48.895\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M48.074-18.41v-25.253\"\u002F>\u003Cpath stroke=\"none\" d=\"m48.074-46.263-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003Cg transform=\"translate(3.733 -78.799)\">\u003Cpath d=\"M50.825 48.895L48.983 44.564Q48.915 44.421 48.734 44.387Q48.553 44.353 48.289 44.353L48.289 44.072L50.258 44.072L50.258 44.353Q49.704 44.353 49.704 44.527Q49.708 44.537 49.709 44.547Q49.711 44.558 49.711 44.564L51.232 48.140L52.664 44.752Q52.688 44.715 52.688 44.660Q52.688 44.500 52.538 44.426Q52.387 44.353 52.203 44.353L52.203 44.072L53.751 44.072L53.751 44.353Q53.491 44.353 53.281 44.447Q53.071 44.541 52.975 44.752L51.219 48.895Q51.167 48.994 51.072 48.994L50.972 48.994Q50.873 48.994 50.825 48.895\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M29.18-18.41C-3.14-39.35-3.14 28.937 26.998 9.412\"\u002F>\u003Cpath stroke=\"none\" d=\"m29.18 7.998-4.622.516 2.44.898-.178 2.593\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(-110.345 -52.31)\">\u003Cpath d=\"M49.175 47.271L49.175 44.564Q49.175 44.353 48.453 44.353L48.453 44.072L50.586 44.072L50.586 44.353Q49.865 44.353 49.865 44.564L49.865 47.251Q49.865 47.668 50.002 47.996Q50.138 48.324 50.420 48.519Q50.702 48.714 51.126 48.714Q51.430 48.714 51.699 48.604Q51.967 48.495 52.167 48.295Q52.367 48.095 52.478 47.829Q52.589 47.562 52.589 47.251L52.589 44.752Q52.589 44.353 51.868 44.353L51.868 44.072L53.591 44.072L53.591 44.353Q52.869 44.353 52.869 44.752L52.869 47.271Q52.869 47.733 52.632 48.133Q52.394 48.533 51.991 48.763Q51.588 48.994 51.126 48.994Q50.641 48.994 50.188 48.775Q49.735 48.557 49.455 48.162Q49.175 47.767 49.175 47.271\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(-110.345 -52.31)\">\u003Cpath d=\"M59.116 50.604Q58.566 50.204 58.195 49.649Q57.824 49.093 57.643 48.447Q57.462 47.801 57.462 47.104Q57.462 46.591 57.562 46.096Q57.663 45.600 57.868 45.149Q58.073 44.698 58.386 44.306Q58.699 43.915 59.116 43.611Q59.126 43.607 59.133 43.606Q59.140 43.604 59.150 43.604L59.218 43.604Q59.253 43.604 59.275 43.628Q59.297 43.652 59.297 43.689Q59.297 43.734 59.270 43.751Q58.921 44.052 58.668 44.436Q58.415 44.821 58.263 45.262Q58.111 45.703 58.039 46.159Q57.967 46.615 57.967 47.104Q57.967 48.105 58.277 48.992Q58.586 49.879 59.270 50.464Q59.297 50.481 59.297 50.525Q59.297 50.563 59.275 50.587Q59.253 50.611 59.218 50.611L59.150 50.611Q59.143 50.607 59.135 50.606Q59.126 50.604 59.116 50.604M61.905 48.854L60.172 48.854L60.172 48.574Q60.398 48.574 60.546 48.540Q60.695 48.505 60.695 48.365L60.695 46.116L60.107 46.116L60.107 45.836L60.695 45.836L60.695 45.019Q60.695 44.701 60.873 44.453Q61.050 44.206 61.341 44.065Q61.631 43.925 61.943 43.925Q62.199 43.925 62.402 44.067Q62.606 44.209 62.606 44.452Q62.606 44.588 62.506 44.687Q62.407 44.787 62.271 44.787Q62.134 44.787 62.035 44.687Q61.936 44.588 61.936 44.452Q61.936 44.271 62.076 44.178Q61.997 44.151 61.898 44.151Q61.690 44.151 61.536 44.284Q61.382 44.417 61.302 44.621Q61.221 44.824 61.221 45.033L61.221 45.836L62.110 45.836L62.110 46.116L61.249 46.116L61.249 48.365Q61.249 48.574 61.905 48.574L61.905 48.854M64.335 48.854L62.599 48.854L62.599 48.574Q62.828 48.574 62.976 48.540Q63.125 48.505 63.125 48.365L63.125 46.516Q63.125 46.246 63.017 46.185Q62.910 46.123 62.599 46.123L62.599 45.843L63.628 45.768L63.628 46.475Q63.757 46.167 64 45.968Q64.243 45.768 64.561 45.768Q64.779 45.768 64.950 45.892Q65.121 46.017 65.121 46.229Q65.121 46.366 65.022 46.465Q64.923 46.564 64.790 46.564Q64.653 46.564 64.554 46.465Q64.455 46.366 64.455 46.229Q64.455 46.089 64.554 45.990Q64.263 45.990 64.063 46.186Q63.863 46.383 63.771 46.677Q63.679 46.971 63.679 47.251L63.679 48.365Q63.679 48.574 64.335 48.574L64.335 48.854M65.665 47.371Q65.665 47.029 65.800 46.730Q65.935 46.431 66.174 46.207Q66.413 45.983 66.731 45.858Q67.049 45.733 67.380 45.733Q67.825 45.733 68.225 45.949Q68.625 46.164 68.859 46.542Q69.093 46.919 69.093 47.371Q69.093 47.712 68.951 47.996Q68.809 48.280 68.565 48.487Q68.320 48.693 68.011 48.808Q67.702 48.922 67.380 48.922Q66.950 48.922 66.548 48.721Q66.147 48.519 65.906 48.167Q65.665 47.815 65.665 47.371M67.380 48.673Q67.982 48.673 68.206 48.295Q68.430 47.917 68.430 47.285Q68.430 46.673 68.196 46.314Q67.962 45.956 67.380 45.956Q66.328 45.956 66.328 47.285Q66.328 47.917 66.553 48.295Q66.779 48.673 67.380 48.673M71.369 48.854L69.735 48.854L69.735 48.574Q69.964 48.574 70.113 48.540Q70.262 48.505 70.262 48.365L70.262 46.516Q70.262 46.246 70.154 46.185Q70.047 46.123 69.735 46.123L69.735 45.843L70.795 45.768L70.795 46.417Q70.966 46.109 71.270 45.938Q71.574 45.768 71.920 45.768Q72.319 45.768 72.596 45.908Q72.873 46.048 72.959 46.396Q73.126 46.103 73.425 45.935Q73.724 45.768 74.069 45.768Q74.575 45.768 74.859 45.991Q75.143 46.215 75.143 46.711L75.143 48.365Q75.143 48.502 75.291 48.538Q75.440 48.574 75.666 48.574L75.666 48.854L74.035 48.854L74.035 48.574Q74.261 48.574 74.411 48.538Q74.562 48.502 74.562 48.365L74.562 46.725Q74.562 46.390 74.442 46.190Q74.322 45.990 74.008 45.990Q73.738 45.990 73.504 46.126Q73.270 46.263 73.131 46.497Q72.993 46.731 72.993 47.005L72.993 48.365Q72.993 48.502 73.141 48.538Q73.290 48.574 73.516 48.574L73.516 48.854L71.885 48.854L71.885 48.574Q72.114 48.574 72.263 48.540Q72.412 48.505 72.412 48.365L72.412 46.725Q72.412 46.390 72.292 46.190Q72.172 45.990 71.858 45.990Q71.588 45.990 71.354 46.126Q71.120 46.263 70.981 46.497Q70.843 46.731 70.843 47.005L70.843 48.365Q70.843 48.502 70.993 48.538Q71.144 48.574 71.369 48.574\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(-110.345 -52.31)\">\u003Cpath d=\"M80.647 48.854L79.013 48.854L79.013 48.574Q79.242 48.574 79.391 48.540Q79.540 48.505 79.540 48.365L79.540 44.746Q79.540 44.476 79.432 44.414Q79.324 44.353 79.013 44.353L79.013 44.072L80.093 43.997L80.093 46.383Q80.199 46.198 80.377 46.056Q80.555 45.915 80.763 45.841Q80.972 45.768 81.197 45.768Q81.703 45.768 81.987 45.991Q82.271 46.215 82.271 46.711L82.271 48.365Q82.271 48.502 82.419 48.538Q82.568 48.574 82.794 48.574L82.794 48.854L81.163 48.854L81.163 48.574Q81.392 48.574 81.541 48.540Q81.690 48.505 81.690 48.365L81.690 46.725Q81.690 46.390 81.570 46.190Q81.450 45.990 81.136 45.990Q80.866 45.990 80.632 46.126Q80.398 46.263 80.259 46.497Q80.121 46.731 80.121 47.005L80.121 48.365Q80.121 48.502 80.271 48.538Q80.422 48.574 80.647 48.574L80.647 48.854M85.511 50.604Q84.961 50.204 84.590 49.649Q84.219 49.093 84.038 48.447Q83.857 47.801 83.857 47.104Q83.857 46.591 83.957 46.096Q84.058 45.600 84.263 45.149Q84.468 44.698 84.781 44.306Q85.094 43.915 85.511 43.611Q85.521 43.607 85.528 43.606Q85.535 43.604 85.545 43.604L85.613 43.604Q85.648 43.604 85.670 43.628Q85.692 43.652 85.692 43.689Q85.692 43.734 85.665 43.751Q85.316 44.052 85.063 44.436Q84.810 44.821 84.658 45.262Q84.506 45.703 84.434 46.159Q84.362 46.615 84.362 47.104Q84.362 48.105 84.672 48.992Q84.981 49.879 85.665 50.464Q85.692 50.481 85.692 50.525Q85.692 50.563 85.670 50.587Q85.648 50.611 85.613 50.611L85.545 50.611Q85.538 50.607 85.530 50.606Q85.521 50.604 85.511 50.604M87.028 48.013L87.028 46.116L86.389 46.116L86.389 45.894Q86.707 45.894 86.924 45.684Q87.141 45.474 87.242 45.164Q87.343 44.855 87.343 44.547L87.609 44.547L87.609 45.836L88.686 45.836L88.686 46.116L87.609 46.116L87.609 48Q87.609 48.276 87.714 48.475Q87.818 48.673 88.078 48.673Q88.235 48.673 88.341 48.569Q88.447 48.464 88.496 48.311Q88.546 48.157 88.546 48L88.546 47.586L88.813 47.586L88.813 48.013Q88.813 48.239 88.713 48.449Q88.614 48.659 88.430 48.791Q88.245 48.922 88.016 48.922Q87.579 48.922 87.304 48.685Q87.028 48.447 87.028 48.013M91.506 47.600L89.448 47.600L89.448 47.097L91.506 47.097L91.506 47.600M95.320 48.854L92.791 48.854L92.791 48.574Q93.758 48.574 93.758 48.365L93.758 44.746Q93.365 44.934 92.743 44.934L92.743 44.653Q93.160 44.653 93.524 44.552Q93.888 44.452 94.145 44.206L94.271 44.206Q94.336 44.223 94.353 44.291L94.353 48.365Q94.353 48.574 95.320 48.574L95.320 48.854M96.612 50.611L96.544 50.611Q96.510 50.611 96.488 50.585Q96.465 50.560 96.465 50.525Q96.465 50.481 96.496 50.464Q96.852 50.160 97.101 49.770Q97.351 49.380 97.503 48.948Q97.655 48.516 97.725 48.047Q97.795 47.579 97.795 47.104Q97.795 46.625 97.725 46.159Q97.655 45.692 97.501 45.257Q97.347 44.821 97.096 44.433Q96.845 44.045 96.496 43.751Q96.465 43.734 96.465 43.689Q96.465 43.655 96.488 43.630Q96.510 43.604 96.544 43.604L96.612 43.604Q96.623 43.604 96.631 43.606Q96.640 43.607 96.650 43.611Q97.193 44.011 97.566 44.564Q97.939 45.118 98.120 45.764Q98.301 46.410 98.301 47.104Q98.301 47.805 98.120 48.452Q97.939 49.100 97.564 49.654Q97.190 50.208 96.650 50.604Q96.640 50.604 96.631 50.606Q96.623 50.607 96.612 50.611M99.733 50.611L99.665 50.611Q99.630 50.611 99.608 50.585Q99.586 50.560 99.586 50.525Q99.586 50.481 99.617 50.464Q99.972 50.160 100.222 49.770Q100.471 49.380 100.623 48.948Q100.776 48.516 100.846 48.047Q100.916 47.579 100.916 47.104Q100.916 46.625 100.846 46.159Q100.776 45.692 100.622 45.257Q100.468 44.821 100.217 44.433Q99.965 44.045 99.617 43.751Q99.586 43.734 99.586 43.689Q99.586 43.655 99.608 43.630Q99.630 43.604 99.665 43.604L99.733 43.604Q99.743 43.604 99.752 43.606Q99.760 43.607 99.771 43.611Q100.314 44.011 100.687 44.564Q101.059 45.118 101.240 45.764Q101.422 46.410 101.422 47.104Q101.422 47.805 101.240 48.452Q101.059 49.100 100.685 49.654Q100.311 50.208 99.771 50.604Q99.760 50.604 99.752 50.606Q99.743 50.607 99.733 50.611\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fsvg>\u003Cfigcaption class=\"tikz-cap\">One RNN cell. The current input \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.5944em;vertical-align:-0.15em;\">\u003C\u002Fspan>\u003Cspan class=\"mord\">\u003Cspan class=\"mord mathbf\">x\u003C\u002Fspan>\u003Cspan class=\"msupsub\">\u003Cspan class=\"vlist-t vlist-t2\">\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.2806em;\">\u003Cspan style=\"top:-2.55em;margin-left:0em;margin-right:0.05em;\">\u003Cspan class=\"pstrut\" style=\"height:2.7em;\">\u003C\u002Fspan>\u003Cspan class=\"sizing reset-size6 size3 mtight\">\u003Cspan class=\"mord mathnormal mtight\">t\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-s\">​\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.15em;\">\u003Cspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> and the previous hidden state \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.9028em;vertical-align:-0.2083em;\">\u003C\u002Fspan>\u003Cspan class=\"mord\">\u003Cspan class=\"mord mathbf\">h\u003C\u002Fspan>\u003Cspan class=\"msupsub\">\u003Cspan class=\"vlist-t vlist-t2\">\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.3011em;\">\u003Cspan style=\"top:-2.55em;margin-left:0em;margin-right:0.05em;\">\u003Cspan class=\"pstrut\" style=\"height:2.7em;\">\u003C\u002Fspan>\u003Cspan class=\"sizing reset-size6 size3 mtight\">\u003Cspan class=\"mord mtight\">\u003Cspan class=\"mord mathnormal mtight\">t\u003C\u002Fspan>\u003Cspan class=\"mbin mtight\">−\u003C\u002Fspan>\u003Cspan class=\"mord mtight\">1\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-s\">​\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.2083em;\">\u003Cspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> combine through \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6861em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathbf\" style=\"margin-right:0.016em;\">W\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> and \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6861em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathbf\">U\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> into the new hidden state \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.8444em;vertical-align:-0.15em;\">\u003C\u002Fspan>\u003Cspan class=\"mord\">\u003Cspan class=\"mord mathbf\">h\u003C\u002Fspan>\u003Cspan class=\"msupsub\">\u003Cspan class=\"vlist-t vlist-t2\">\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.2806em;\">\u003Cspan style=\"top:-2.55em;margin-left:0em;margin-right:0.05em;\">\u003Cspan class=\"pstrut\" style=\"height:2.7em;\">\u003C\u002Fspan>\u003Cspan class=\"sizing reset-size6 size3 mtight\">\u003Cspan class=\"mord mathnormal mtight\">t\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-s\">​\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.15em;\">\u003Cspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>, which both feeds the output \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6389em;vertical-align:-0.1944em;\">\u003C\u002Fspan>\u003Cspan class=\"mord\">\u003Cspan class=\"mord mathbf\" style=\"margin-right:0.016em;\">y\u003C\u002Fspan>\u003Cspan class=\"msupsub\">\u003Cspan class=\"vlist-t vlist-t2\">\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.2806em;\">\u003Cspan style=\"top:-2.55em;margin-left:-0.016em;margin-right:0.05em;\">\u003Cspan class=\"pstrut\" style=\"height:2.7em;\">\u003C\u002Fspan>\u003Cspan class=\"sizing reset-size6 size3 mtight\">\u003Cspan class=\"mord mathnormal mtight\">t\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-s\">​\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.15em;\">\u003Cspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> and is passed forward in time.\u003C\u002Ffigcaption>","\u003Csvg style=\"width:100%;max-width:359.922px;height:auto\" xmlns=\"http:\u002F\u002Fwww.w3.org\u002F2000\u002Fsvg\" viewBox=\"-75 -75 269.942 137.283\">\u003Cg stroke=\"currentColor\" style=\"stroke-miterlimit:10;stroke-width:.4\">\u003Cpath fill=\"none\" d=\"M-27.815 58.813H3.483V36.05h-31.298Z\"\u002F>\u003Cg transform=\"translate(-8.35 2.25)\">\u003Cpath d=\"M-10.364 47.432L-12.039 47.432L-12.039 47.116Q-11.696 47.116-11.390 46.975Q-11.085 46.834-10.870 46.566L-10.043 45.551L-11.111 44.167Q-11.283 43.965-11.465 43.916Q-11.647 43.868-11.995 43.868L-11.995 43.552L-10.096 43.552L-10.096 43.868Q-10.206 43.868-10.301 43.923Q-10.395 43.978-10.395 44.075Q-10.395 44.114-10.364 44.167L-9.657 45.072L-9.116 44.417Q-9.006 44.277-9.006 44.127Q-9.006 44.017-9.074 43.943Q-9.143 43.868-9.248 43.868L-9.248 43.552L-7.578 43.552L-7.578 43.868Q-7.921 43.868-8.229 44.011Q-8.536 44.154-8.747 44.417L-9.477 45.309L-8.294 46.821Q-8.176 46.962-8.046 47.026Q-7.916 47.089-7.785 47.102Q-7.653 47.116-7.416 47.116L-7.416 47.432L-9.323 47.432L-9.323 47.116Q-9.217 47.116-9.116 47.061Q-9.015 47.006-9.015 46.909Q-9.015 46.861-9.041 46.821L-9.854 45.775L-10.500 46.566Q-10.610 46.716-10.610 46.856Q-10.610 46.962-10.538 47.039Q-10.465 47.116-10.364 47.116L-10.364 47.432M-4.436 49.673Q-4.941 49.286-5.311 48.781Q-5.680 48.276-5.924 47.676Q-6.167 47.076-6.282 46.459Q-6.396 45.841-6.396 45.182Q-6.396 44.523-6.282 43.908Q-6.167 43.292-5.928 42.699Q-5.688 42.106-5.315 41.596Q-4.941 41.086-4.436 40.700Q-4.401 40.682-4.379 40.682L-4.300 40.682Q-4.212 40.682-4.212 40.783Q-4.212 40.818-4.247 40.853Q-4.810 41.376-5.155 42.082Q-5.499 42.787-5.647 43.569Q-5.794 44.351-5.794 45.182Q-5.794 45.806-5.715 46.390Q-5.636 46.975-5.458 47.542Q-5.280 48.109-4.981 48.610Q-4.682 49.111-4.247 49.519Q-4.212 49.555-4.212 49.594Q-4.212 49.691-4.300 49.691L-4.379 49.691Q-4.401 49.691-4.436 49.673M0.209 47.432L-2.823 47.432L-2.823 47.116Q-1.672 47.116-1.672 46.821L-1.672 42.097Q-2.160 42.330-2.880 42.330L-2.880 42.014Q-1.751 42.014-1.188 41.438L-1.043 41.438Q-1.008 41.438-0.975 41.471Q-0.942 41.504-0.942 41.539L-0.942 46.821Q-0.942 47.116 0.209 47.116L0.209 47.432M1.606 49.691L1.523 49.691Q1.435 49.691 1.435 49.594Q1.435 49.555 1.470 49.519Q2.305 48.746 2.666 47.612Q3.026 46.478 3.026 45.182Q3.026 44.562 2.947 43.971Q2.868 43.380 2.688 42.822Q2.507 42.264 2.206 41.756Q1.905 41.249 1.470 40.853Q1.435 40.818 1.435 40.783Q1.435 40.682 1.523 40.682L1.606 40.682Q1.624 40.682 1.659 40.700Q2.165 41.082 2.538 41.596Q2.912 42.110 3.149 42.688Q3.386 43.266 3.503 43.894Q3.619 44.523 3.619 45.182Q3.619 45.841 3.503 46.472Q3.386 47.102 3.147 47.687Q2.907 48.271 2.536 48.781Q2.165 49.291 1.659 49.673Q1.624 49.691 1.606 49.691\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-27.815 4.753H3.483V-18.01h-31.298Z\"\u002F>\u003Cg transform=\"translate(-8.48 -51.81)\">\u003Cpath d=\"M-9.780 47.432L-11.867 47.432L-11.867 47.116Q-11.560 47.116-11.368 47.063Q-11.177 47.010-11.177 46.821L-11.177 42.106Q-11.177 41.864-11.248 41.756Q-11.318 41.649-11.452 41.625Q-11.586 41.600-11.867 41.600L-11.867 41.284L-10.500 41.187L-10.500 44.237Q-10.303 43.881-9.949 43.668Q-9.595 43.455-9.195 43.455Q-7.916 43.455-7.916 44.668L-7.916 46.821Q-7.916 47.010-7.725 47.063Q-7.534 47.116-7.227 47.116L-7.227 47.432L-9.314 47.432L-9.314 47.116Q-9.002 47.116-8.811 47.063Q-8.620 47.010-8.620 46.821L-8.620 44.703Q-8.620 44.444-8.664 44.222Q-8.708 44-8.853 43.857Q-8.998 43.714-9.257 43.714Q-9.600 43.714-9.881 43.903Q-10.162 44.092-10.318 44.404Q-10.474 44.716-10.474 45.063L-10.474 46.821Q-10.474 47.010-10.281 47.063Q-10.087 47.116-9.780 47.116L-9.780 47.432M-4.177 49.673Q-4.682 49.286-5.051 48.781Q-5.420 48.276-5.664 47.676Q-5.908 47.076-6.022 46.459Q-6.137 45.841-6.137 45.182Q-6.137 44.523-6.022 43.908Q-5.908 43.292-5.669 42.699Q-5.429 42.106-5.056 41.596Q-4.682 41.086-4.177 40.700Q-4.142 40.682-4.120 40.682L-4.041 40.682Q-3.953 40.682-3.953 40.783Q-3.953 40.818-3.988 40.853Q-4.550 41.376-4.895 42.082Q-5.240 42.787-5.387 43.569Q-5.535 44.351-5.535 45.182Q-5.535 45.806-5.456 46.390Q-5.376 46.975-5.198 47.542Q-5.020 48.109-4.722 48.610Q-4.423 49.111-3.988 49.519Q-3.953 49.555-3.953 49.594Q-3.953 49.691-4.041 49.691L-4.120 49.691Q-4.142 49.691-4.177 49.673M0.468 47.432L-2.564 47.432L-2.564 47.116Q-1.413 47.116-1.413 46.821L-1.413 42.097Q-1.900 42.330-2.621 42.330L-2.621 42.014Q-1.492 42.014-0.929 41.438L-0.784 41.438Q-0.749 41.438-0.716 41.471Q-0.683 41.504-0.683 41.539L-0.683 46.821Q-0.683 47.116 0.468 47.116L0.468 47.432M1.866 49.691L1.782 49.691Q1.694 49.691 1.694 49.594Q1.694 49.555 1.730 49.519Q2.564 48.746 2.925 47.612Q3.285 46.478 3.285 45.182Q3.285 44.562 3.206 43.971Q3.127 43.380 2.947 42.822Q2.767 42.264 2.466 41.756Q2.165 41.249 1.730 40.853Q1.694 40.818 1.694 40.783Q1.694 40.682 1.782 40.682L1.866 40.682Q1.883 40.682 1.918 40.700Q2.424 41.082 2.797 41.596Q3.171 42.110 3.408 42.688Q3.646 43.266 3.762 43.894Q3.878 44.523 3.878 45.182Q3.878 45.841 3.762 46.472Q3.646 47.102 3.406 47.687Q3.167 48.271 2.795 48.781Q2.424 49.291 1.918 49.673Q1.883 49.691 1.866 49.691\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-27.815-49.308H3.483V-72.07h-31.298Z\"\u002F>\u003Cg transform=\"translate(-8.35 -105.87)\">\u003Cpath d=\"M-11.489 48.909Q-11.322 49.014-11.142 49.014Q-10.817 49.014-10.580 48.770Q-10.342 48.526-10.193 48.179L-9.890 47.432L-11.256 44.167Q-11.349 43.969-11.511 43.919Q-11.674 43.868-11.977 43.868L-11.977 43.552L-10.052 43.552L-10.052 43.868Q-10.544 43.868-10.544 44.083Q-10.544 44.105-10.527 44.167L-9.520 46.557L-8.620 44.417Q-8.584 44.334-8.584 44.228Q-8.584 44.066-8.705 43.967Q-8.826 43.868-8.989 43.868L-8.989 43.552L-7.486 43.552L-7.486 43.868Q-7.771 43.868-7.985 44.011Q-8.198 44.154-8.303 44.417L-9.890 48.179Q-10.079 48.632-10.393 48.955Q-10.707 49.278-11.142 49.278Q-11.467 49.278-11.727 49.058Q-11.986 48.838-11.986 48.513Q-11.986 48.346-11.867 48.232Q-11.749 48.118-11.582 48.118Q-11.467 48.118-11.377 48.168Q-11.287 48.219-11.237 48.307Q-11.186 48.394-11.186 48.513Q-11.186 48.658-11.267 48.768Q-11.349 48.878-11.489 48.909M-4.436 49.673Q-4.941 49.286-5.311 48.781Q-5.680 48.276-5.924 47.676Q-6.167 47.076-6.282 46.459Q-6.396 45.841-6.396 45.182Q-6.396 44.523-6.282 43.908Q-6.167 43.292-5.928 42.699Q-5.688 42.106-5.315 41.596Q-4.941 41.086-4.436 40.700Q-4.401 40.682-4.379 40.682L-4.300 40.682Q-4.212 40.682-4.212 40.783Q-4.212 40.818-4.247 40.853Q-4.810 41.376-5.155 42.082Q-5.499 42.787-5.647 43.569Q-5.794 44.351-5.794 45.182Q-5.794 45.806-5.715 46.390Q-5.636 46.975-5.458 47.542Q-5.280 48.109-4.981 48.610Q-4.682 49.111-4.247 49.519Q-4.212 49.555-4.212 49.594Q-4.212 49.691-4.300 49.691L-4.379 49.691Q-4.401 49.691-4.436 49.673M0.209 47.432L-2.823 47.432L-2.823 47.116Q-1.672 47.116-1.672 46.821L-1.672 42.097Q-2.160 42.330-2.880 42.330L-2.880 42.014Q-1.751 42.014-1.188 41.438L-1.043 41.438Q-1.008 41.438-0.975 41.471Q-0.942 41.504-0.942 41.539L-0.942 46.821Q-0.942 47.116 0.209 47.116L0.209 47.432M1.606 49.691L1.523 49.691Q1.435 49.691 1.435 49.594Q1.435 49.555 1.470 49.519Q2.305 48.746 2.666 47.612Q3.026 46.478 3.026 45.182Q3.026 44.562 2.947 43.971Q2.868 43.380 2.688 42.822Q2.507 42.264 2.206 41.756Q1.905 41.249 1.470 40.853Q1.435 40.818 1.435 40.783Q1.435 40.682 1.523 40.682L1.606 40.682Q1.624 40.682 1.659 40.700Q2.165 41.082 2.538 41.596Q2.912 42.110 3.149 42.688Q3.386 43.266 3.503 43.894Q3.619 44.523 3.619 45.182Q3.619 45.841 3.503 46.472Q3.386 47.102 3.147 47.687Q2.907 48.271 2.536 48.781Q2.165 49.291 1.659 49.673Q1.624 49.691 1.606 49.691\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-12.166 35.85V7.753\"\u002F>\u003Cpath stroke=\"none\" d=\"m-12.166 5.153-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003Cg transform=\"translate(3.733 -24.538)\">\u003Cpath d=\"M-9.773 47.473L-11.325 43.142Q-11.387 42.999-11.549 42.965Q-11.711 42.931-11.964 42.931L-11.964 42.650L-10.037 42.650L-10.037 42.931Q-10.618 42.931-10.618 43.105Q-10.618 43.125-10.611 43.142L-9.387 46.578L-8.280 43.491L-8.406 43.142Q-8.464 42.999-8.630 42.965Q-8.796 42.931-9.045 42.931L-9.045 42.650L-7.118 42.650L-7.118 42.931Q-7.699 42.931-7.699 43.105L-7.699 43.142L-6.468 46.578L-5.313 43.330Q-5.299 43.289-5.299 43.265Q-5.299 43.091-5.492 43.011Q-5.686 42.931-5.894 42.931L-5.894 42.650L-4.318 42.650L-4.318 42.931Q-4.568 42.931-4.759 43.025Q-4.951 43.119-5.026 43.330L-6.509 47.473Q-6.543 47.572-6.643 47.572L-6.721 47.572Q-6.820 47.572-6.861 47.473L-8.140 43.884L-9.421 47.473Q-9.438 47.517-9.476 47.545Q-9.514 47.572-9.562 47.572L-9.640 47.572Q-9.732 47.572-9.773 47.473\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-12.166-18.41v-28.098\"\u002F>\u003Cpath stroke=\"none\" d=\"m-12.166-49.108-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003Cg transform=\"translate(3.733 -78.799)\">\u003Cpath d=\"M-9.415 47.473L-11.257 43.142Q-11.325 42.999-11.506 42.965Q-11.687 42.931-11.951 42.931L-11.951 42.650L-9.982 42.650L-9.982 42.931Q-10.536 42.931-10.536 43.105Q-10.532 43.115-10.531 43.125Q-10.529 43.136-10.529 43.142L-9.008 46.718L-7.576 43.330Q-7.552 43.293-7.552 43.238Q-7.552 43.078-7.702 43.004Q-7.853 42.931-8.037 42.931L-8.037 42.650L-6.489 42.650L-6.489 42.931Q-6.749 42.931-6.959 43.025Q-7.169 43.119-7.265 43.330L-9.021 47.473Q-9.073 47.572-9.168 47.572L-9.268 47.572Q-9.367 47.572-9.415 47.473\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M66.08 58.813h31.297V36.05H66.08Z\"\u002F>\u003Cg transform=\"translate(85.543 2.25)\">\u003Cpath d=\"M-10.364 47.432L-12.039 47.432L-12.039 47.116Q-11.696 47.116-11.390 46.975Q-11.085 46.834-10.870 46.566L-10.043 45.551L-11.111 44.167Q-11.283 43.965-11.465 43.916Q-11.647 43.868-11.995 43.868L-11.995 43.552L-10.096 43.552L-10.096 43.868Q-10.206 43.868-10.301 43.923Q-10.395 43.978-10.395 44.075Q-10.395 44.114-10.364 44.167L-9.657 45.072L-9.116 44.417Q-9.006 44.277-9.006 44.127Q-9.006 44.017-9.074 43.943Q-9.143 43.868-9.248 43.868L-9.248 43.552L-7.578 43.552L-7.578 43.868Q-7.921 43.868-8.229 44.011Q-8.536 44.154-8.747 44.417L-9.477 45.309L-8.294 46.821Q-8.176 46.962-8.046 47.026Q-7.916 47.089-7.785 47.102Q-7.653 47.116-7.416 47.116L-7.416 47.432L-9.323 47.432L-9.323 47.116Q-9.217 47.116-9.116 47.061Q-9.015 47.006-9.015 46.909Q-9.015 46.861-9.041 46.821L-9.854 45.775L-10.500 46.566Q-10.610 46.716-10.610 46.856Q-10.610 46.962-10.538 47.039Q-10.465 47.116-10.364 47.116L-10.364 47.432M-4.436 49.673Q-4.941 49.286-5.311 48.781Q-5.680 48.276-5.924 47.676Q-6.167 47.076-6.282 46.459Q-6.396 45.841-6.396 45.182Q-6.396 44.523-6.282 43.908Q-6.167 43.292-5.928 42.699Q-5.688 42.106-5.315 41.596Q-4.941 41.086-4.436 40.700Q-4.401 40.682-4.379 40.682L-4.300 40.682Q-4.212 40.682-4.212 40.783Q-4.212 40.818-4.247 40.853Q-4.810 41.376-5.155 42.082Q-5.499 42.787-5.647 43.569Q-5.794 44.351-5.794 45.182Q-5.794 45.806-5.715 46.390Q-5.636 46.975-5.458 47.542Q-5.280 48.109-4.981 48.610Q-4.682 49.111-4.247 49.519Q-4.212 49.555-4.212 49.594Q-4.212 49.691-4.300 49.691L-4.379 49.691Q-4.401 49.691-4.436 49.673M0.209 47.432L-3.241 47.432L-3.241 47.199Q-3.241 47.186-3.210 47.155L-1.755 45.578Q-1.290 45.081-1.037 44.776Q-0.784 44.470-0.593 44.059Q-0.402 43.648-0.402 43.209Q-0.402 42.620-0.725 42.187Q-1.048 41.754-1.628 41.754Q-1.892 41.754-2.138 41.864Q-2.384 41.974-2.560 42.161Q-2.735 42.348-2.832 42.598L-2.753 42.598Q-2.551 42.598-2.408 42.734Q-2.265 42.870-2.265 43.086Q-2.265 43.292-2.408 43.431Q-2.551 43.569-2.753 43.569Q-2.955 43.569-3.098 43.426Q-3.241 43.284-3.241 43.086Q-3.241 42.624-3.003 42.251Q-2.766 41.877-2.366 41.658Q-1.966 41.438-1.518 41.438Q-0.995 41.438-0.540 41.653Q-0.085 41.869 0.187 42.268Q0.459 42.668 0.459 43.209Q0.459 43.604 0.288 43.958Q0.117 44.312-0.149 44.591Q-0.415 44.870-0.865 45.255Q-1.316 45.639-1.395 45.714L-2.419 46.676L-1.602 46.676Q-0.951 46.676-0.514 46.665Q-0.077 46.654-0.046 46.632Q0.024 46.549 0.079 46.309Q0.134 46.070 0.174 45.802L0.459 45.802L0.209 47.432M1.606 49.691L1.523 49.691Q1.435 49.691 1.435 49.594Q1.435 49.555 1.470 49.519Q2.305 48.746 2.666 47.612Q3.026 46.478 3.026 45.182Q3.026 44.562 2.947 43.971Q2.868 43.380 2.688 42.822Q2.507 42.264 2.206 41.756Q1.905 41.249 1.470 40.853Q1.435 40.818 1.435 40.783Q1.435 40.682 1.523 40.682L1.606 40.682Q1.624 40.682 1.659 40.700Q2.165 41.082 2.538 41.596Q2.912 42.110 3.149 42.688Q3.386 43.266 3.503 43.894Q3.619 44.523 3.619 45.182Q3.619 45.841 3.503 46.472Q3.386 47.102 3.147 47.687Q2.907 48.271 2.536 48.781Q2.165 49.291 1.659 49.673Q1.624 49.691 1.606 49.691\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M66.08 4.753h31.297V-18.01H66.08Z\"\u002F>\u003Cg transform=\"translate(85.415 -51.81)\">\u003Cpath d=\"M-9.780 47.432L-11.867 47.432L-11.867 47.116Q-11.560 47.116-11.368 47.063Q-11.177 47.010-11.177 46.821L-11.177 42.106Q-11.177 41.864-11.248 41.756Q-11.318 41.649-11.452 41.625Q-11.586 41.600-11.867 41.600L-11.867 41.284L-10.500 41.187L-10.500 44.237Q-10.303 43.881-9.949 43.668Q-9.595 43.455-9.195 43.455Q-7.916 43.455-7.916 44.668L-7.916 46.821Q-7.916 47.010-7.725 47.063Q-7.534 47.116-7.227 47.116L-7.227 47.432L-9.314 47.432L-9.314 47.116Q-9.002 47.116-8.811 47.063Q-8.620 47.010-8.620 46.821L-8.620 44.703Q-8.620 44.444-8.664 44.222Q-8.708 44-8.853 43.857Q-8.998 43.714-9.257 43.714Q-9.600 43.714-9.881 43.903Q-10.162 44.092-10.318 44.404Q-10.474 44.716-10.474 45.063L-10.474 46.821Q-10.474 47.010-10.281 47.063Q-10.087 47.116-9.780 47.116L-9.780 47.432M-4.177 49.673Q-4.682 49.286-5.051 48.781Q-5.420 48.276-5.664 47.676Q-5.908 47.076-6.022 46.459Q-6.137 45.841-6.137 45.182Q-6.137 44.523-6.022 43.908Q-5.908 43.292-5.669 42.699Q-5.429 42.106-5.056 41.596Q-4.682 41.086-4.177 40.700Q-4.142 40.682-4.120 40.682L-4.041 40.682Q-3.953 40.682-3.953 40.783Q-3.953 40.818-3.988 40.853Q-4.550 41.376-4.895 42.082Q-5.240 42.787-5.387 43.569Q-5.535 44.351-5.535 45.182Q-5.535 45.806-5.456 46.390Q-5.376 46.975-5.198 47.542Q-5.020 48.109-4.722 48.610Q-4.423 49.111-3.988 49.519Q-3.953 49.555-3.953 49.594Q-3.953 49.691-4.041 49.691L-4.120 49.691Q-4.142 49.691-4.177 49.673M0.468 47.432L-2.981 47.432L-2.981 47.199Q-2.981 47.186-2.951 47.155L-1.496 45.578Q-1.030 45.081-0.778 44.776Q-0.525 44.470-0.334 44.059Q-0.143 43.648-0.143 43.209Q-0.143 42.620-0.466 42.187Q-0.789 41.754-1.369 41.754Q-1.632 41.754-1.878 41.864Q-2.124 41.974-2.300 42.161Q-2.476 42.348-2.573 42.598L-2.494 42.598Q-2.291 42.598-2.149 42.734Q-2.006 42.870-2.006 43.086Q-2.006 43.292-2.149 43.431Q-2.291 43.569-2.494 43.569Q-2.696 43.569-2.839 43.426Q-2.981 43.284-2.981 43.086Q-2.981 42.624-2.744 42.251Q-2.507 41.877-2.107 41.658Q-1.707 41.438-1.259 41.438Q-0.736 41.438-0.281 41.653Q0.174 41.869 0.446 42.268Q0.719 42.668 0.719 43.209Q0.719 43.604 0.547 43.958Q0.376 44.312 0.110 44.591Q-0.156 44.870-0.606 45.255Q-1.057 45.639-1.136 45.714L-2.160 46.676L-1.342 46.676Q-0.692 46.676-0.255 46.665Q0.183 46.654 0.213 46.632Q0.284 46.549 0.339 46.309Q0.394 46.070 0.433 45.802L0.719 45.802L0.468 47.432M1.866 49.691L1.782 49.691Q1.694 49.691 1.694 49.594Q1.694 49.555 1.730 49.519Q2.564 48.746 2.925 47.612Q3.285 46.478 3.285 45.182Q3.285 44.562 3.206 43.971Q3.127 43.380 2.947 42.822Q2.767 42.264 2.466 41.756Q2.165 41.249 1.730 40.853Q1.694 40.818 1.694 40.783Q1.694 40.682 1.782 40.682L1.866 40.682Q1.883 40.682 1.918 40.700Q2.424 41.082 2.797 41.596Q3.171 42.110 3.408 42.688Q3.646 43.266 3.762 43.894Q3.878 44.523 3.878 45.182Q3.878 45.841 3.762 46.472Q3.646 47.102 3.406 47.687Q3.167 48.271 2.795 48.781Q2.424 49.291 1.918 49.673Q1.883 49.691 1.866 49.691\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M66.08-49.308h31.297V-72.07H66.08Z\"\u002F>\u003Cg transform=\"translate(85.543 -105.87)\">\u003Cpath d=\"M-11.489 48.909Q-11.322 49.014-11.142 49.014Q-10.817 49.014-10.580 48.770Q-10.342 48.526-10.193 48.179L-9.890 47.432L-11.256 44.167Q-11.349 43.969-11.511 43.919Q-11.674 43.868-11.977 43.868L-11.977 43.552L-10.052 43.552L-10.052 43.868Q-10.544 43.868-10.544 44.083Q-10.544 44.105-10.527 44.167L-9.520 46.557L-8.620 44.417Q-8.584 44.334-8.584 44.228Q-8.584 44.066-8.705 43.967Q-8.826 43.868-8.989 43.868L-8.989 43.552L-7.486 43.552L-7.486 43.868Q-7.771 43.868-7.985 44.011Q-8.198 44.154-8.303 44.417L-9.890 48.179Q-10.079 48.632-10.393 48.955Q-10.707 49.278-11.142 49.278Q-11.467 49.278-11.727 49.058Q-11.986 48.838-11.986 48.513Q-11.986 48.346-11.867 48.232Q-11.749 48.118-11.582 48.118Q-11.467 48.118-11.377 48.168Q-11.287 48.219-11.237 48.307Q-11.186 48.394-11.186 48.513Q-11.186 48.658-11.267 48.768Q-11.349 48.878-11.489 48.909M-4.436 49.673Q-4.941 49.286-5.311 48.781Q-5.680 48.276-5.924 47.676Q-6.167 47.076-6.282 46.459Q-6.396 45.841-6.396 45.182Q-6.396 44.523-6.282 43.908Q-6.167 43.292-5.928 42.699Q-5.688 42.106-5.315 41.596Q-4.941 41.086-4.436 40.700Q-4.401 40.682-4.379 40.682L-4.300 40.682Q-4.212 40.682-4.212 40.783Q-4.212 40.818-4.247 40.853Q-4.810 41.376-5.155 42.082Q-5.499 42.787-5.647 43.569Q-5.794 44.351-5.794 45.182Q-5.794 45.806-5.715 46.390Q-5.636 46.975-5.458 47.542Q-5.280 48.109-4.981 48.610Q-4.682 49.111-4.247 49.519Q-4.212 49.555-4.212 49.594Q-4.212 49.691-4.300 49.691L-4.379 49.691Q-4.401 49.691-4.436 49.673M0.209 47.432L-3.241 47.432L-3.241 47.199Q-3.241 47.186-3.210 47.155L-1.755 45.578Q-1.290 45.081-1.037 44.776Q-0.784 44.470-0.593 44.059Q-0.402 43.648-0.402 43.209Q-0.402 42.620-0.725 42.187Q-1.048 41.754-1.628 41.754Q-1.892 41.754-2.138 41.864Q-2.384 41.974-2.560 42.161Q-2.735 42.348-2.832 42.598L-2.753 42.598Q-2.551 42.598-2.408 42.734Q-2.265 42.870-2.265 43.086Q-2.265 43.292-2.408 43.431Q-2.551 43.569-2.753 43.569Q-2.955 43.569-3.098 43.426Q-3.241 43.284-3.241 43.086Q-3.241 42.624-3.003 42.251Q-2.766 41.877-2.366 41.658Q-1.966 41.438-1.518 41.438Q-0.995 41.438-0.540 41.653Q-0.085 41.869 0.187 42.268Q0.459 42.668 0.459 43.209Q0.459 43.604 0.288 43.958Q0.117 44.312-0.149 44.591Q-0.415 44.870-0.865 45.255Q-1.316 45.639-1.395 45.714L-2.419 46.676L-1.602 46.676Q-0.951 46.676-0.514 46.665Q-0.077 46.654-0.046 46.632Q0.024 46.549 0.079 46.309Q0.134 46.070 0.174 45.802L0.459 45.802L0.209 47.432M1.606 49.691L1.523 49.691Q1.435 49.691 1.435 49.594Q1.435 49.555 1.470 49.519Q2.305 48.746 2.666 47.612Q3.026 46.478 3.026 45.182Q3.026 44.562 2.947 43.971Q2.868 43.380 2.688 42.822Q2.507 42.264 2.206 41.756Q1.905 41.249 1.470 40.853Q1.435 40.818 1.435 40.783Q1.435 40.682 1.523 40.682L1.606 40.682Q1.624 40.682 1.659 40.700Q2.165 41.082 2.538 41.596Q2.912 42.110 3.149 42.688Q3.386 43.266 3.503 43.894Q3.619 44.523 3.619 45.182Q3.619 45.841 3.503 46.472Q3.386 47.102 3.147 47.687Q2.907 48.271 2.536 48.781Q2.165 49.291 1.659 49.673Q1.624 49.691 1.606 49.691\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M81.728 35.85V7.753\"\u002F>\u003Cpath stroke=\"none\" d=\"m81.728 5.153-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003Cg transform=\"translate(97.627 -24.538)\">\u003Cpath d=\"M-9.773 47.473L-11.325 43.142Q-11.387 42.999-11.549 42.965Q-11.711 42.931-11.964 42.931L-11.964 42.650L-10.037 42.650L-10.037 42.931Q-10.618 42.931-10.618 43.105Q-10.618 43.125-10.611 43.142L-9.387 46.578L-8.280 43.491L-8.406 43.142Q-8.464 42.999-8.630 42.965Q-8.796 42.931-9.045 42.931L-9.045 42.650L-7.118 42.650L-7.118 42.931Q-7.699 42.931-7.699 43.105L-7.699 43.142L-6.468 46.578L-5.313 43.330Q-5.299 43.289-5.299 43.265Q-5.299 43.091-5.492 43.011Q-5.686 42.931-5.894 42.931L-5.894 42.650L-4.318 42.650L-4.318 42.931Q-4.568 42.931-4.759 43.025Q-4.951 43.119-5.026 43.330L-6.509 47.473Q-6.543 47.572-6.643 47.572L-6.721 47.572Q-6.820 47.572-6.861 47.473L-8.140 43.884L-9.421 47.473Q-9.438 47.517-9.476 47.545Q-9.514 47.572-9.562 47.572L-9.640 47.572Q-9.732 47.572-9.773 47.473\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M81.728-18.41v-28.098\"\u002F>\u003Cpath stroke=\"none\" d=\"m81.728-49.108-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003Cg transform=\"translate(97.627 -78.799)\">\u003Cpath d=\"M-9.415 47.473L-11.257 43.142Q-11.325 42.999-11.506 42.965Q-11.687 42.931-11.951 42.931L-11.951 42.650L-9.982 42.650L-9.982 42.931Q-10.536 42.931-10.536 43.105Q-10.532 43.115-10.531 43.125Q-10.529 43.136-10.529 43.142L-9.008 46.718L-7.576 43.330Q-7.552 43.293-7.552 43.238Q-7.552 43.078-7.702 43.004Q-7.853 42.931-8.037 42.931L-8.037 42.650L-6.489 42.650L-6.489 42.931Q-6.749 42.931-6.959 43.025Q-7.169 43.119-7.265 43.330L-9.021 47.473Q-9.073 47.572-9.168 47.572L-9.268 47.572Q-9.367 47.572-9.415 47.473\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M159.974 58.813h31.298V36.05h-31.298Z\"\u002F>\u003Cg transform=\"translate(179.438 2.25)\">\u003Cpath d=\"M-10.364 47.432L-12.039 47.432L-12.039 47.116Q-11.696 47.116-11.390 46.975Q-11.085 46.834-10.870 46.566L-10.043 45.551L-11.111 44.167Q-11.283 43.965-11.465 43.916Q-11.647 43.868-11.995 43.868L-11.995 43.552L-10.096 43.552L-10.096 43.868Q-10.206 43.868-10.301 43.923Q-10.395 43.978-10.395 44.075Q-10.395 44.114-10.364 44.167L-9.657 45.072L-9.116 44.417Q-9.006 44.277-9.006 44.127Q-9.006 44.017-9.074 43.943Q-9.143 43.868-9.248 43.868L-9.248 43.552L-7.578 43.552L-7.578 43.868Q-7.921 43.868-8.229 44.011Q-8.536 44.154-8.747 44.417L-9.477 45.309L-8.294 46.821Q-8.176 46.962-8.046 47.026Q-7.916 47.089-7.785 47.102Q-7.653 47.116-7.416 47.116L-7.416 47.432L-9.323 47.432L-9.323 47.116Q-9.217 47.116-9.116 47.061Q-9.015 47.006-9.015 46.909Q-9.015 46.861-9.041 46.821L-9.854 45.775L-10.500 46.566Q-10.610 46.716-10.610 46.856Q-10.610 46.962-10.538 47.039Q-10.465 47.116-10.364 47.116L-10.364 47.432M-4.436 49.673Q-4.941 49.286-5.311 48.781Q-5.680 48.276-5.924 47.676Q-6.167 47.076-6.282 46.459Q-6.396 45.841-6.396 45.182Q-6.396 44.523-6.282 43.908Q-6.167 43.292-5.928 42.699Q-5.688 42.106-5.315 41.596Q-4.941 41.086-4.436 40.700Q-4.401 40.682-4.379 40.682L-4.300 40.682Q-4.212 40.682-4.212 40.783Q-4.212 40.818-4.247 40.853Q-4.810 41.376-5.155 42.082Q-5.499 42.787-5.647 43.569Q-5.794 44.351-5.794 45.182Q-5.794 45.806-5.715 46.390Q-5.636 46.975-5.458 47.542Q-5.280 48.109-4.981 48.610Q-4.682 49.111-4.247 49.519Q-4.212 49.555-4.212 49.594Q-4.212 49.691-4.300 49.691L-4.379 49.691Q-4.401 49.691-4.436 49.673M-2.797 46.711L-2.841 46.711Q-2.639 47.028-2.252 47.186Q-1.865 47.344-1.439 47.344Q-0.903 47.344-0.663 46.909Q-0.424 46.474-0.424 45.894Q-0.424 45.314-0.670 44.874Q-0.916 44.435-1.448 44.435L-2.067 44.435Q-2.094 44.435-2.127 44.406Q-2.160 44.378-2.160 44.356L-2.160 44.255Q-2.160 44.224-2.131 44.200Q-2.103 44.176-2.067 44.176L-1.549 44.136Q-1.083 44.136-0.837 43.664Q-0.591 43.191-0.591 42.673Q-0.591 42.246-0.804 41.972Q-1.017 41.697-1.439 41.697Q-1.782 41.697-2.107 41.827Q-2.432 41.956-2.617 42.211L-2.590 42.211Q-2.388 42.211-2.252 42.352Q-2.116 42.493-2.116 42.690Q-2.116 42.888-2.250 43.022Q-2.384 43.156-2.582 43.156Q-2.784 43.156-2.922 43.022Q-3.061 42.888-3.061 42.690Q-3.061 42.101-2.557 41.770Q-2.054 41.438-1.439 41.438Q-1.061 41.438-0.659 41.578Q-0.257 41.719 0.011 41.998Q0.279 42.277 0.279 42.673Q0.279 43.222-0.074 43.659Q-0.428 44.097-0.969 44.281Q-0.578 44.360-0.233 44.584Q0.112 44.808 0.323 45.149Q0.534 45.490 0.534 45.885Q0.534 46.267 0.372 46.590Q0.209 46.913-0.083 47.149Q-0.375 47.384-0.723 47.507Q-1.070 47.630-1.439 47.630Q-1.887 47.630-2.318 47.469Q-2.749 47.309-3.030 46.982Q-3.311 46.654-3.311 46.197Q-3.311 45.982-3.164 45.839Q-3.017 45.696-2.797 45.696Q-2.586 45.696-2.441 45.841Q-2.296 45.986-2.296 46.197Q-2.296 46.408-2.443 46.560Q-2.590 46.711-2.797 46.711M1.606 49.691L1.523 49.691Q1.435 49.691 1.435 49.594Q1.435 49.555 1.470 49.519Q2.305 48.746 2.666 47.612Q3.026 46.478 3.026 45.182Q3.026 44.562 2.947 43.971Q2.868 43.380 2.688 42.822Q2.507 42.264 2.206 41.756Q1.905 41.249 1.470 40.853Q1.435 40.818 1.435 40.783Q1.435 40.682 1.523 40.682L1.606 40.682Q1.624 40.682 1.659 40.700Q2.165 41.082 2.538 41.596Q2.912 42.110 3.149 42.688Q3.386 43.266 3.503 43.894Q3.619 44.523 3.619 45.182Q3.619 45.841 3.503 46.472Q3.386 47.102 3.147 47.687Q2.907 48.271 2.536 48.781Q2.165 49.291 1.659 49.673Q1.624 49.691 1.606 49.691\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M159.974 4.753h31.298V-18.01h-31.298Z\"\u002F>\u003Cg transform=\"translate(179.31 -51.81)\">\u003Cpath d=\"M-9.780 47.432L-11.867 47.432L-11.867 47.116Q-11.560 47.116-11.368 47.063Q-11.177 47.010-11.177 46.821L-11.177 42.106Q-11.177 41.864-11.248 41.756Q-11.318 41.649-11.452 41.625Q-11.586 41.600-11.867 41.600L-11.867 41.284L-10.500 41.187L-10.500 44.237Q-10.303 43.881-9.949 43.668Q-9.595 43.455-9.195 43.455Q-7.916 43.455-7.916 44.668L-7.916 46.821Q-7.916 47.010-7.725 47.063Q-7.534 47.116-7.227 47.116L-7.227 47.432L-9.314 47.432L-9.314 47.116Q-9.002 47.116-8.811 47.063Q-8.620 47.010-8.620 46.821L-8.620 44.703Q-8.620 44.444-8.664 44.222Q-8.708 44-8.853 43.857Q-8.998 43.714-9.257 43.714Q-9.600 43.714-9.881 43.903Q-10.162 44.092-10.318 44.404Q-10.474 44.716-10.474 45.063L-10.474 46.821Q-10.474 47.010-10.281 47.063Q-10.087 47.116-9.780 47.116L-9.780 47.432M-4.177 49.673Q-4.682 49.286-5.051 48.781Q-5.420 48.276-5.664 47.676Q-5.908 47.076-6.022 46.459Q-6.137 45.841-6.137 45.182Q-6.137 44.523-6.022 43.908Q-5.908 43.292-5.669 42.699Q-5.429 42.106-5.056 41.596Q-4.682 41.086-4.177 40.700Q-4.142 40.682-4.120 40.682L-4.041 40.682Q-3.953 40.682-3.953 40.783Q-3.953 40.818-3.988 40.853Q-4.550 41.376-4.895 42.082Q-5.240 42.787-5.387 43.569Q-5.535 44.351-5.535 45.182Q-5.535 45.806-5.456 46.390Q-5.376 46.975-5.198 47.542Q-5.020 48.109-4.722 48.610Q-4.423 49.111-3.988 49.519Q-3.953 49.555-3.953 49.594Q-3.953 49.691-4.041 49.691L-4.120 49.691Q-4.142 49.691-4.177 49.673M-2.538 46.711L-2.582 46.711Q-2.379 47.028-1.993 47.186Q-1.606 47.344-1.180 47.344Q-0.644 47.344-0.404 46.909Q-0.165 46.474-0.165 45.894Q-0.165 45.314-0.411 44.874Q-0.657 44.435-1.188 44.435L-1.808 44.435Q-1.834 44.435-1.867 44.406Q-1.900 44.378-1.900 44.356L-1.900 44.255Q-1.900 44.224-1.872 44.200Q-1.843 44.176-1.808 44.176L-1.290 44.136Q-0.824 44.136-0.578 43.664Q-0.332 43.191-0.332 42.673Q-0.332 42.246-0.545 41.972Q-0.758 41.697-1.180 41.697Q-1.522 41.697-1.848 41.827Q-2.173 41.956-2.357 42.211L-2.331 42.211Q-2.129 42.211-1.993 42.352Q-1.856 42.493-1.856 42.690Q-1.856 42.888-1.990 43.022Q-2.124 43.156-2.322 43.156Q-2.524 43.156-2.663 43.022Q-2.801 42.888-2.801 42.690Q-2.801 42.101-2.298 41.770Q-1.795 41.438-1.180 41.438Q-0.802 41.438-0.400 41.578Q0.002 41.719 0.271 41.998Q0.539 42.277 0.539 42.673Q0.539 43.222 0.185 43.659Q-0.169 44.097-0.709 44.281Q-0.318 44.360 0.027 44.584Q0.372 44.808 0.583 45.149Q0.793 45.490 0.793 45.885Q0.793 46.267 0.631 46.590Q0.468 46.913 0.176 47.149Q-0.116 47.384-0.463 47.507Q-0.811 47.630-1.180 47.630Q-1.628 47.630-2.059 47.469Q-2.489 47.309-2.770 46.982Q-3.052 46.654-3.052 46.197Q-3.052 45.982-2.905 45.839Q-2.757 45.696-2.538 45.696Q-2.327 45.696-2.182 45.841Q-2.037 45.986-2.037 46.197Q-2.037 46.408-2.184 46.560Q-2.331 46.711-2.538 46.711M1.866 49.691L1.782 49.691Q1.694 49.691 1.694 49.594Q1.694 49.555 1.730 49.519Q2.564 48.746 2.925 47.612Q3.285 46.478 3.285 45.182Q3.285 44.562 3.206 43.971Q3.127 43.380 2.947 42.822Q2.767 42.264 2.466 41.756Q2.165 41.249 1.730 40.853Q1.694 40.818 1.694 40.783Q1.694 40.682 1.782 40.682L1.866 40.682Q1.883 40.682 1.918 40.700Q2.424 41.082 2.797 41.596Q3.171 42.110 3.408 42.688Q3.646 43.266 3.762 43.894Q3.878 44.523 3.878 45.182Q3.878 45.841 3.762 46.472Q3.646 47.102 3.406 47.687Q3.167 48.271 2.795 48.781Q2.424 49.291 1.918 49.673Q1.883 49.691 1.866 49.691\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M159.974-49.308h31.298V-72.07h-31.298Z\"\u002F>\u003Cg transform=\"translate(179.438 -105.87)\">\u003Cpath d=\"M-11.489 48.909Q-11.322 49.014-11.142 49.014Q-10.817 49.014-10.580 48.770Q-10.342 48.526-10.193 48.179L-9.890 47.432L-11.256 44.167Q-11.349 43.969-11.511 43.919Q-11.674 43.868-11.977 43.868L-11.977 43.552L-10.052 43.552L-10.052 43.868Q-10.544 43.868-10.544 44.083Q-10.544 44.105-10.527 44.167L-9.520 46.557L-8.620 44.417Q-8.584 44.334-8.584 44.228Q-8.584 44.066-8.705 43.967Q-8.826 43.868-8.989 43.868L-8.989 43.552L-7.486 43.552L-7.486 43.868Q-7.771 43.868-7.985 44.011Q-8.198 44.154-8.303 44.417L-9.890 48.179Q-10.079 48.632-10.393 48.955Q-10.707 49.278-11.142 49.278Q-11.467 49.278-11.727 49.058Q-11.986 48.838-11.986 48.513Q-11.986 48.346-11.867 48.232Q-11.749 48.118-11.582 48.118Q-11.467 48.118-11.377 48.168Q-11.287 48.219-11.237 48.307Q-11.186 48.394-11.186 48.513Q-11.186 48.658-11.267 48.768Q-11.349 48.878-11.489 48.909M-4.436 49.673Q-4.941 49.286-5.311 48.781Q-5.680 48.276-5.924 47.676Q-6.167 47.076-6.282 46.459Q-6.396 45.841-6.396 45.182Q-6.396 44.523-6.282 43.908Q-6.167 43.292-5.928 42.699Q-5.688 42.106-5.315 41.596Q-4.941 41.086-4.436 40.700Q-4.401 40.682-4.379 40.682L-4.300 40.682Q-4.212 40.682-4.212 40.783Q-4.212 40.818-4.247 40.853Q-4.810 41.376-5.155 42.082Q-5.499 42.787-5.647 43.569Q-5.794 44.351-5.794 45.182Q-5.794 45.806-5.715 46.390Q-5.636 46.975-5.458 47.542Q-5.280 48.109-4.981 48.610Q-4.682 49.111-4.247 49.519Q-4.212 49.555-4.212 49.594Q-4.212 49.691-4.300 49.691L-4.379 49.691Q-4.401 49.691-4.436 49.673M-2.797 46.711L-2.841 46.711Q-2.639 47.028-2.252 47.186Q-1.865 47.344-1.439 47.344Q-0.903 47.344-0.663 46.909Q-0.424 46.474-0.424 45.894Q-0.424 45.314-0.670 44.874Q-0.916 44.435-1.448 44.435L-2.067 44.435Q-2.094 44.435-2.127 44.406Q-2.160 44.378-2.160 44.356L-2.160 44.255Q-2.160 44.224-2.131 44.200Q-2.103 44.176-2.067 44.176L-1.549 44.136Q-1.083 44.136-0.837 43.664Q-0.591 43.191-0.591 42.673Q-0.591 42.246-0.804 41.972Q-1.017 41.697-1.439 41.697Q-1.782 41.697-2.107 41.827Q-2.432 41.956-2.617 42.211L-2.590 42.211Q-2.388 42.211-2.252 42.352Q-2.116 42.493-2.116 42.690Q-2.116 42.888-2.250 43.022Q-2.384 43.156-2.582 43.156Q-2.784 43.156-2.922 43.022Q-3.061 42.888-3.061 42.690Q-3.061 42.101-2.557 41.770Q-2.054 41.438-1.439 41.438Q-1.061 41.438-0.659 41.578Q-0.257 41.719 0.011 41.998Q0.279 42.277 0.279 42.673Q0.279 43.222-0.074 43.659Q-0.428 44.097-0.969 44.281Q-0.578 44.360-0.233 44.584Q0.112 44.808 0.323 45.149Q0.534 45.490 0.534 45.885Q0.534 46.267 0.372 46.590Q0.209 46.913-0.083 47.149Q-0.375 47.384-0.723 47.507Q-1.070 47.630-1.439 47.630Q-1.887 47.630-2.318 47.469Q-2.749 47.309-3.030 46.982Q-3.311 46.654-3.311 46.197Q-3.311 45.982-3.164 45.839Q-3.017 45.696-2.797 45.696Q-2.586 45.696-2.441 45.841Q-2.296 45.986-2.296 46.197Q-2.296 46.408-2.443 46.560Q-2.590 46.711-2.797 46.711M1.606 49.691L1.523 49.691Q1.435 49.691 1.435 49.594Q1.435 49.555 1.470 49.519Q2.305 48.746 2.666 47.612Q3.026 46.478 3.026 45.182Q3.026 44.562 2.947 43.971Q2.868 43.380 2.688 42.822Q2.507 42.264 2.206 41.756Q1.905 41.249 1.470 40.853Q1.435 40.818 1.435 40.783Q1.435 40.682 1.523 40.682L1.606 40.682Q1.624 40.682 1.659 40.700Q2.165 41.082 2.538 41.596Q2.912 42.110 3.149 42.688Q3.386 43.266 3.503 43.894Q3.619 44.523 3.619 45.182Q3.619 45.841 3.503 46.472Q3.386 47.102 3.147 47.687Q2.907 48.271 2.536 48.781Q2.165 49.291 1.659 49.673Q1.624 49.691 1.606 49.691\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M175.623 35.85V7.753\"\u002F>\u003Cpath stroke=\"none\" d=\"m175.623 5.153-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003Cg transform=\"translate(191.521 -24.538)\">\u003Cpath d=\"M-9.773 47.473L-11.325 43.142Q-11.387 42.999-11.549 42.965Q-11.711 42.931-11.964 42.931L-11.964 42.650L-10.037 42.650L-10.037 42.931Q-10.618 42.931-10.618 43.105Q-10.618 43.125-10.611 43.142L-9.387 46.578L-8.280 43.491L-8.406 43.142Q-8.464 42.999-8.630 42.965Q-8.796 42.931-9.045 42.931L-9.045 42.650L-7.118 42.650L-7.118 42.931Q-7.699 42.931-7.699 43.105L-7.699 43.142L-6.468 46.578L-5.313 43.330Q-5.299 43.289-5.299 43.265Q-5.299 43.091-5.492 43.011Q-5.686 42.931-5.894 42.931L-5.894 42.650L-4.318 42.650L-4.318 42.931Q-4.568 42.931-4.759 43.025Q-4.951 43.119-5.026 43.330L-6.509 47.473Q-6.543 47.572-6.643 47.572L-6.721 47.572Q-6.820 47.572-6.861 47.473L-8.140 43.884L-9.421 47.473Q-9.438 47.517-9.476 47.545Q-9.514 47.572-9.562 47.572L-9.640 47.572Q-9.732 47.572-9.773 47.473\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M175.623-18.41v-28.098\"\u002F>\u003Cpath stroke=\"none\" d=\"m175.623-49.108-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003Cg transform=\"translate(191.521 -78.799)\">\u003Cpath d=\"M-9.415 47.473L-11.257 43.142Q-11.325 42.999-11.506 42.965Q-11.687 42.931-11.951 42.931L-11.951 42.650L-9.982 42.650L-9.982 42.931Q-10.536 42.931-10.536 43.105Q-10.532 43.115-10.531 43.125Q-10.529 43.136-10.529 43.142L-9.008 46.718L-7.576 43.330Q-7.552 43.293-7.552 43.238Q-7.552 43.078-7.702 43.004Q-7.853 42.931-8.037 42.931L-8.037 42.650L-6.489 42.650L-6.489 42.931Q-6.749 42.931-6.959 43.025Q-7.169 43.119-7.265 43.330L-9.021 47.473Q-9.073 47.572-9.168 47.572L-9.268 47.572Q-9.367 47.572-9.415 47.473\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg transform=\"translate(-39.925 -52.31)\">\u003Cpath d=\"M-10.170 47.432L-11.804 47.432L-11.804 47.152Q-11.575 47.152-11.426 47.118Q-11.277 47.083-11.277 46.943L-11.277 43.324Q-11.277 43.054-11.385 42.992Q-11.493 42.931-11.804 42.931L-11.804 42.650L-10.724 42.575L-10.724 44.961Q-10.618 44.776-10.440 44.634Q-10.262 44.493-10.054 44.419Q-9.845 44.346-9.620 44.346Q-9.114 44.346-8.830 44.569Q-8.546 44.793-8.546 45.289L-8.546 46.943Q-8.546 47.080-8.398 47.116Q-8.249 47.152-8.023 47.152L-8.023 47.432L-9.654 47.432L-9.654 47.152Q-9.425 47.152-9.276 47.118Q-9.127 47.083-9.127 46.943L-9.127 45.303Q-9.127 44.968-9.247 44.768Q-9.367 44.568-9.681 44.568Q-9.951 44.568-10.185 44.704Q-10.419 44.841-10.558 45.075Q-10.696 45.309-10.696 45.583L-10.696 46.943Q-10.696 47.080-10.546 47.116Q-10.395 47.152-10.170 47.152L-10.170 47.432M-5.306 49.182Q-5.856 48.782-6.227 48.227Q-6.598 47.671-6.779 47.025Q-6.960 46.379-6.960 45.682Q-6.960 45.169-6.860 44.674Q-6.759 44.178-6.554 43.727Q-6.349 43.276-6.036 42.884Q-5.723 42.493-5.306 42.189Q-5.296 42.185-5.289 42.184Q-5.282 42.182-5.272 42.182L-5.204 42.182Q-5.169 42.182-5.147 42.206Q-5.125 42.230-5.125 42.267Q-5.125 42.312-5.152 42.329Q-5.501 42.630-5.754 43.014Q-6.007 43.399-6.159 43.840Q-6.311 44.281-6.383 44.737Q-6.455 45.193-6.455 45.682Q-6.455 46.683-6.145 47.570Q-5.836 48.457-5.152 49.042Q-5.125 49.059-5.125 49.103Q-5.125 49.141-5.147 49.165Q-5.169 49.189-5.204 49.189L-5.272 49.189Q-5.279 49.185-5.287 49.184Q-5.296 49.182-5.306 49.182M-2.640 47.572Q-3.276 47.572-3.640 47.227Q-4.004 46.882-4.139 46.357Q-4.274 45.832-4.274 45.207Q-4.274 44.182-3.918 43.483Q-3.563 42.784-2.640 42.784Q-1.714 42.784-1.362 43.483Q-1.010 44.182-1.010 45.207Q-1.010 45.832-1.145 46.357Q-1.280 46.882-1.642 47.227Q-2.004 47.572-2.640 47.572M-2.640 47.347Q-2.203 47.347-1.989 46.972Q-1.775 46.598-1.726 46.131Q-1.676 45.665-1.676 45.087Q-1.676 44.534-1.726 44.106Q-1.775 43.679-1.987 43.344Q-2.199 43.009-2.640 43.009Q-2.982 43.009-3.185 43.216Q-3.389 43.423-3.476 43.735Q-3.563 44.048-3.585 44.364Q-3.607 44.681-3.607 45.087Q-3.607 45.504-3.585 45.846Q-3.563 46.188-3.474 46.536Q-3.385 46.885-3.180 47.116Q-2.975 47.347-2.640 47.347M-0.012 49.189L-0.080 49.189Q-0.114 49.189-0.136 49.163Q-0.159 49.138-0.159 49.103Q-0.159 49.059-0.128 49.042Q0.228 48.738 0.477 48.348Q0.727 47.958 0.879 47.526Q1.031 47.094 1.101 46.625Q1.171 46.157 1.171 45.682Q1.171 45.203 1.101 44.737Q1.031 44.270 0.877 43.835Q0.723 43.399 0.472 43.011Q0.221 42.623-0.128 42.329Q-0.159 42.312-0.159 42.267Q-0.159 42.233-0.136 42.208Q-0.114 42.182-0.080 42.182L-0.012 42.182Q-0.001 42.182 0.007 42.184Q0.016 42.185 0.026 42.189Q0.569 42.589 0.942 43.142Q1.314 43.696 1.496 44.342Q1.677 44.988 1.677 45.682Q1.677 46.383 1.496 47.030Q1.314 47.678 0.940 48.232Q0.566 48.786 0.026 49.182Q0.016 49.182 0.007 49.184Q-0.001 49.185-0.012 49.189\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-65.203-6.628h34.388\"\u002F>\u003Cpath stroke=\"none\" d=\"m-28.215-6.628-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M3.883-6.628H63.08\"\u002F>\u003Cpath stroke=\"none\" d=\"m65.68-6.628-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003Cg transform=\"translate(43.996 -57.793)\">\u003Cpath d=\"M-11.065 45.849L-11.065 43.142Q-11.065 42.931-11.787 42.931L-11.787 42.650L-9.654 42.650L-9.654 42.931Q-10.375 42.931-10.375 43.142L-10.375 45.829Q-10.375 46.246-10.238 46.574Q-10.102 46.902-9.820 47.097Q-9.538 47.292-9.114 47.292Q-8.810 47.292-8.541 47.182Q-8.273 47.073-8.073 46.873Q-7.873 46.673-7.762 46.407Q-7.651 46.140-7.651 45.829L-7.651 43.330Q-7.651 42.931-8.372 42.931L-8.372 42.650L-6.649 42.650L-6.649 42.931Q-7.371 42.931-7.371 43.330L-7.371 45.849Q-7.371 46.311-7.608 46.711Q-7.846 47.111-8.249 47.341Q-8.652 47.572-9.114 47.572Q-9.599 47.572-10.052 47.353Q-10.505 47.135-10.785 46.740Q-11.065 46.345-11.065 45.849\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M97.777-6.628h59.197\"\u002F>\u003Cpath stroke=\"none\" d=\"m159.574-6.628-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003Cg transform=\"translate(137.89 -57.793)\">\u003Cpath d=\"M-11.065 45.849L-11.065 43.142Q-11.065 42.931-11.787 42.931L-11.787 42.650L-9.654 42.650L-9.654 42.931Q-10.375 42.931-10.375 43.142L-10.375 45.829Q-10.375 46.246-10.238 46.574Q-10.102 46.902-9.820 47.097Q-9.538 47.292-9.114 47.292Q-8.810 47.292-8.541 47.182Q-8.273 47.073-8.073 46.873Q-7.873 46.673-7.762 46.407Q-7.651 46.140-7.651 45.829L-7.651 43.330Q-7.651 42.931-8.372 42.931L-8.372 42.650L-6.649 42.650L-6.649 42.931Q-7.371 42.931-7.371 43.330L-7.371 45.849Q-7.371 46.311-7.608 46.711Q-7.846 47.111-8.249 47.341Q-8.652 47.572-9.114 47.572Q-9.599 47.572-10.052 47.353Q-10.505 47.135-10.785 46.740Q-11.065 46.345-11.065 45.849\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fsvg>\u003Cfigcaption class=\"tikz-cap\">The same RNN unrolled across three steps. Layers are recomputed at every step, but the weights \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6861em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathbf\">U\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>, \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6861em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathbf\" style=\"margin-right:0.016em;\">V\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>, \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6861em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathbf\" style=\"margin-right:0.016em;\">W\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> are shared; the hidden state is the one wire that carries context from each step to the next.\u003C\u002Ffigcaption>","\u003Csvg style=\"width:100%;max-width:417.023px;height:auto\" xmlns=\"http:\u002F\u002Fwww.w3.org\u002F2000\u002Fsvg\" viewBox=\"-75 -75 312.767 148.664\">\u003Cg stroke=\"currentColor\" style=\"stroke-miterlimit:10;stroke-width:.4\">\u003Cpath fill=\"none\" d=\"M-30.14 70.194h25.608V53.122H-30.14Z\"\u002F>\u003Cg transform=\"translate(-4.201 2.392)\">\u003Cpath d=\"M-16.847 61.720L-16.847 60.147Q-16.847 60.120-16.822 60.094Q-16.796 60.069-16.769 60.069L-16.656 60.069Q-16.628 60.069-16.605 60.096Q-16.581 60.123-16.581 60.147Q-16.581 60.492-16.449 60.756Q-16.317 61.019-16.088 61.188Q-15.859 61.357-15.557 61.438Q-15.254 61.518-14.913 61.518Q-14.646 61.518-14.410 61.390Q-14.174 61.262-14.029 61.039Q-13.884 60.817-13.884 60.551Q-13.884 60.328-13.990 60.132Q-14.096 59.935-14.277 59.800Q-14.458 59.665-14.684 59.614L-15.712 59.382Q-16.023 59.310-16.283 59.124Q-16.543 58.937-16.695 58.666Q-16.847 58.394-16.847 58.079Q-16.847 57.693-16.634 57.386Q-16.420 57.078-16.073 56.907Q-15.726 56.736-15.347 56.736Q-15.118 56.736-14.889 56.789Q-14.660 56.842-14.461 56.950Q-14.263 57.057-14.109 57.221L-13.815 56.781Q-13.792 56.736-13.751 56.736L-13.703 56.736Q-13.672 56.736-13.650 56.762Q-13.628 56.787-13.628 56.815L-13.628 58.390Q-13.628 58.411-13.651 58.438Q-13.675 58.466-13.703 58.466L-13.815 58.466Q-13.877 58.466-13.891 58.390Q-13.932 57.977-14.113 57.657Q-14.294 57.338-14.605 57.163Q-14.916 56.989-15.347 56.989Q-15.596 56.989-15.836 57.100Q-16.075 57.211-16.225 57.409Q-16.376 57.608-16.376 57.871Q-16.376 58.083-16.268 58.264Q-16.160 58.445-15.984 58.565Q-15.808 58.684-15.600 58.725L-14.571 58.954Q-14.253 59.026-13.986 59.231Q-13.720 59.436-13.568 59.730Q-13.416 60.024-13.416 60.356Q-13.416 60.749-13.621 61.084Q-13.826 61.419-14.171 61.608Q-14.516 61.798-14.913 61.798Q-15.333 61.798-15.712 61.685Q-16.092 61.573-16.362 61.323L-16.656 61.757Q-16.683 61.798-16.721 61.798L-16.769 61.798Q-16.796 61.798-16.822 61.773Q-16.847 61.747-16.847 61.720M-12.647 60.175Q-12.647 59.833-12.512 59.534Q-12.377 59.235-12.137 59.011Q-11.898 58.787-11.580 58.662Q-11.262 58.537-10.931 58.537Q-10.486 58.537-10.086 58.753Q-9.687 58.968-9.452 59.346Q-9.218 59.723-9.218 60.175Q-9.218 60.516-9.360 60.800Q-9.502 61.084-9.746 61.291Q-9.991 61.497-10.300 61.612Q-10.609 61.726-10.931 61.726Q-11.361 61.726-11.763 61.525Q-12.165 61.323-12.406 60.971Q-12.647 60.619-12.647 60.175M-10.931 61.477Q-10.329 61.477-10.105 61.099Q-9.881 60.721-9.881 60.089Q-9.881 59.477-10.116 59.118Q-10.350 58.760-10.931 58.760Q-11.983 58.760-11.983 60.089Q-11.983 60.721-11.758 61.099Q-11.532 61.477-10.931 61.477\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-30.14 30.36h25.608V7.597H-30.14Z\"\u002F>\u003Cg transform=\"translate(-2.57 -39.554)\">\u003Cpath d=\"M-14.950 61.658L-17.037 61.658L-17.037 61.342Q-16.730 61.342-16.538 61.289Q-16.347 61.236-16.347 61.047L-16.347 56.332Q-16.347 56.090-16.418 55.982Q-16.488 55.875-16.622 55.851Q-16.756 55.826-17.037 55.826L-17.037 55.510L-15.670 55.413L-15.670 58.463Q-15.473 58.107-15.119 57.894Q-14.765 57.681-14.365 57.681Q-13.086 57.681-13.086 58.894L-13.086 61.047Q-13.086 61.236-12.895 61.289Q-12.704 61.342-12.397 61.342L-12.397 61.658L-14.484 61.658L-14.484 61.342Q-14.172 61.342-13.981 61.289Q-13.790 61.236-13.790 61.047L-13.790 58.929Q-13.790 58.670-13.834 58.448Q-13.878 58.226-14.023 58.083Q-14.168 57.940-14.427 57.940Q-14.770 57.940-15.051 58.129Q-15.332 58.318-15.488 58.630Q-15.644 58.942-15.644 59.289L-15.644 61.047Q-15.644 61.236-15.451 61.289Q-15.257 61.342-14.950 61.342\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-34.427-15.165H-.245v-17.071h-34.182Z\"\u002F>\u003Cg transform=\"translate(-13.758 -82.928)\">\u003Cpath d=\"M-17.022 61.651L-17.022 60.588Q-17.022 60.564-16.994 60.537Q-16.967 60.510-16.943 60.510L-16.834 60.510Q-16.769 60.510-16.755 60.568Q-16.659 61.002-16.413 61.253Q-16.167 61.504-15.753 61.504Q-15.412 61.504-15.159 61.371Q-14.906 61.238-14.906 60.930Q-14.906 60.773-15 60.658Q-15.094 60.544-15.232 60.475Q-15.371 60.407-15.538 60.369L-16.119 60.270Q-16.475 60.202-16.748 59.981Q-17.022 59.761-17.022 59.419Q-17.022 59.170-16.910 58.995Q-16.799 58.821-16.613 58.722Q-16.427 58.623-16.211 58.580Q-15.996 58.537-15.753 58.537Q-15.340 58.537-15.060 58.719L-14.844 58.544Q-14.834 58.541-14.827 58.539Q-14.820 58.537-14.810 58.537L-14.759 58.537Q-14.732 58.537-14.708 58.561Q-14.684 58.585-14.684 58.613L-14.684 59.460Q-14.684 59.481-14.708 59.508Q-14.732 59.535-14.759 59.535L-14.872 59.535Q-14.899 59.535-14.925 59.510Q-14.950 59.484-14.950 59.460Q-14.950 59.224-15.056 59.060Q-15.162 58.896-15.345 58.814Q-15.528 58.732-15.760 58.732Q-16.088 58.732-16.345 58.835Q-16.601 58.937-16.601 59.214Q-16.601 59.409-16.418 59.518Q-16.235 59.628-16.006 59.669L-15.432 59.775Q-15.186 59.823-14.972 59.951Q-14.759 60.079-14.622 60.282Q-14.485 60.486-14.485 60.735Q-14.485 61.248-14.851 61.487Q-15.217 61.726-15.753 61.726Q-16.249 61.726-16.581 61.432L-16.847 61.706Q-16.868 61.726-16.895 61.726L-16.943 61.726Q-16.967 61.726-16.994 61.699Q-17.022 61.672-17.022 61.651M-13.898 60.175Q-13.898 59.833-13.763 59.534Q-13.628 59.235-13.388 59.011Q-13.149 58.787-12.831 58.662Q-12.513 58.537-12.182 58.537Q-11.737 58.537-11.337 58.753Q-10.938 58.968-10.703 59.346Q-10.469 59.723-10.469 60.175Q-10.469 60.516-10.611 60.800Q-10.753 61.084-10.997 61.291Q-11.242 61.497-11.551 61.612Q-11.860 61.726-12.182 61.726Q-12.612 61.726-13.014 61.525Q-13.416 61.323-13.657 60.971Q-13.898 60.619-13.898 60.175M-12.182 61.477Q-11.580 61.477-11.356 61.099Q-11.132 60.721-11.132 60.089Q-11.132 59.477-11.367 59.118Q-11.601 58.760-12.182 58.760Q-13.234 58.760-13.234 60.089Q-13.234 60.721-13.009 61.099Q-12.783 61.477-12.182 61.477M-8.077 61.658L-9.810 61.658L-9.810 61.378Q-9.584 61.378-9.435 61.344Q-9.287 61.309-9.287 61.169L-9.287 58.920L-9.875 58.920L-9.875 58.640L-9.287 58.640L-9.287 57.823Q-9.287 57.505-9.109 57.257Q-8.931 57.010-8.641 56.869Q-8.350 56.729-8.039 56.729Q-7.783 56.729-7.579 56.871Q-7.376 57.013-7.376 57.256Q-7.376 57.392-7.475 57.491Q-7.574 57.591-7.711 57.591Q-7.848 57.591-7.947 57.491Q-8.046 57.392-8.046 57.256Q-8.046 57.075-7.906 56.982Q-7.984 56.955-8.084 56.955Q-8.292 56.955-8.446 57.088Q-8.600 57.221-8.680 57.425Q-8.760 57.628-8.760 57.837L-8.760 58.640L-7.872 58.640L-7.872 58.920L-8.733 58.920L-8.733 61.169Q-8.733 61.378-8.077 61.378L-8.077 61.658M-6.870 60.817L-6.870 58.920L-7.509 58.920L-7.509 58.698Q-7.191 58.698-6.974 58.488Q-6.757 58.278-6.657 57.968Q-6.556 57.659-6.556 57.351L-6.289 57.351L-6.289 58.640L-5.212 58.640L-5.212 58.920L-6.289 58.920L-6.289 60.804Q-6.289 61.080-6.185 61.279Q-6.081 61.477-5.821 61.477Q-5.664 61.477-5.558 61.373Q-5.452 61.268-5.402 61.115Q-5.353 60.961-5.353 60.804L-5.353 60.390L-5.086 60.390L-5.086 60.817Q-5.086 61.043-5.185 61.253Q-5.284 61.463-5.469 61.595Q-5.653 61.726-5.882 61.726Q-6.320 61.726-6.595 61.489Q-6.870 61.251-6.870 60.817M-2.594 61.658L-4.228 61.658L-4.228 61.378Q-3.999 61.378-3.850 61.344Q-3.702 61.309-3.702 61.169L-3.702 59.320Q-3.702 59.050-3.809 58.989Q-3.917 58.927-4.228 58.927L-4.228 58.647L-3.169 58.572L-3.169 59.221Q-2.998 58.913-2.693 58.742Q-2.389 58.572-2.044 58.572Q-1.644 58.572-1.367 58.712Q-1.090 58.852-1.005 59.200Q-0.837 58.907-0.538 58.739Q-0.239 58.572 0.106 58.572Q0.612 58.572 0.895 58.795Q1.179 59.019 1.179 59.515L1.179 61.169Q1.179 61.306 1.328 61.342Q1.477 61.378 1.702 61.378L1.702 61.658L0.072 61.658L0.072 61.378Q0.297 61.378 0.448 61.342Q0.598 61.306 0.598 61.169L0.598 59.529Q0.598 59.194 0.478 58.994Q0.359 58.794 0.044 58.794Q-0.226 58.794-0.460 58.930Q-0.694 59.067-0.832 59.301Q-0.971 59.535-0.971 59.809L-0.971 61.169Q-0.971 61.306-0.822 61.342Q-0.673 61.378-0.448 61.378L-0.448 61.658L-2.078 61.658L-2.078 61.378Q-1.849 61.378-1.701 61.344Q-1.552 61.309-1.552 61.169L-1.552 59.529Q-1.552 59.194-1.671 58.994Q-1.791 58.794-2.106 58.794Q-2.376 58.794-2.610 58.930Q-2.844 59.067-2.982 59.301Q-3.121 59.535-3.121 59.809L-3.121 61.169Q-3.121 61.306-2.970 61.342Q-2.820 61.378-2.594 61.378L-2.594 61.658M2.348 60.930Q2.348 60.598 2.572 60.371Q2.796 60.144 3.139 60.016Q3.483 59.887 3.855 59.835Q4.228 59.782 4.532 59.782L4.532 59.529Q4.532 59.324 4.424 59.144Q4.317 58.965 4.136 58.862Q3.955 58.760 3.746 58.760Q3.339 58.760 3.103 58.852Q3.192 58.889 3.238 58.973Q3.285 59.057 3.285 59.159Q3.285 59.255 3.238 59.334Q3.192 59.412 3.112 59.457Q3.032 59.501 2.943 59.501Q2.792 59.501 2.692 59.404Q2.591 59.306 2.591 59.159Q2.591 58.537 3.746 58.537Q3.958 58.537 4.207 58.601Q4.457 58.664 4.659 58.783Q4.860 58.903 4.987 59.088Q5.113 59.272 5.113 59.515L5.113 61.091Q5.113 61.207 5.175 61.303Q5.236 61.398 5.349 61.398Q5.458 61.398 5.523 61.304Q5.588 61.210 5.588 61.091L5.588 60.643L5.855 60.643L5.855 61.091Q5.855 61.361 5.628 61.526Q5.400 61.692 5.120 61.692Q4.912 61.692 4.775 61.538Q4.638 61.385 4.614 61.169Q4.467 61.436 4.185 61.581Q3.903 61.726 3.579 61.726Q3.302 61.726 3.018 61.651Q2.734 61.576 2.541 61.397Q2.348 61.217 2.348 60.930M2.963 60.930Q2.963 61.104 3.064 61.234Q3.165 61.364 3.320 61.434Q3.476 61.504 3.640 61.504Q3.859 61.504 4.067 61.407Q4.276 61.309 4.404 61.128Q4.532 60.947 4.532 60.721L4.532 59.993Q4.207 59.993 3.842 60.084Q3.476 60.175 3.220 60.387Q2.963 60.598 2.963 60.930M7.455 61.658L6.132 61.658L6.132 61.378Q6.692 61.378 7.072 60.978L7.786 60.181L6.873 59.132Q6.737 58.985 6.588 58.953Q6.439 58.920 6.173 58.920L6.173 58.640L7.673 58.640L7.673 58.920Q7.482 58.920 7.482 59.054Q7.482 59.084 7.513 59.132L8.107 59.816L8.548 59.320Q8.661 59.190 8.661 59.074Q8.661 59.012 8.623 58.966Q8.586 58.920 8.528 58.920L8.528 58.640L9.844 58.640L9.844 58.920Q9.283 58.920 8.904 59.320L8.282 60.021L9.276 61.169Q9.375 61.268 9.476 61.313Q9.577 61.357 9.688 61.367Q9.799 61.378 9.977 61.378L9.977 61.658L8.483 61.658L8.483 61.378Q8.548 61.378 8.608 61.344Q8.668 61.309 8.668 61.244Q8.668 61.197 8.637 61.169L7.960 60.383L7.427 60.978Q7.314 61.108 7.314 61.224Q7.314 61.289 7.355 61.333Q7.396 61.378 7.455 61.378\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-41.034-54.998H6.362V-72.07h-47.396Z\"\u002F>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(-20.365 -123.442)\">\u003Cpath d=\"M-15.138 60.404L-17.196 60.404L-17.196 59.901L-15.138 59.901L-15.138 60.404M-12.667 61.658L-14.270 61.658L-14.270 61.378Q-14.044 61.378-13.896 61.344Q-13.747 61.309-13.747 61.169L-13.747 57.550Q-13.747 57.280-13.855 57.218Q-13.962 57.157-14.270 57.157L-14.270 56.876L-13.193 56.801L-13.193 61.169Q-13.193 61.306-13.043 61.342Q-12.893 61.378-12.667 61.378L-12.667 61.658M-12.113 60.175Q-12.113 59.833-11.978 59.534Q-11.843 59.235-11.604 59.011Q-11.365 58.787-11.047 58.662Q-10.729 58.537-10.398 58.537Q-9.953 58.537-9.553 58.753Q-9.153 58.968-8.919 59.346Q-8.685 59.723-8.685 60.175Q-8.685 60.516-8.827 60.800Q-8.969 61.084-9.213 61.291Q-9.458 61.497-9.767 61.612Q-10.076 61.726-10.398 61.726Q-10.828 61.726-11.230 61.525Q-11.631 61.323-11.872 60.971Q-12.113 60.619-12.113 60.175M-10.398 61.477Q-9.796 61.477-9.572 61.099Q-9.348 60.721-9.348 60.089Q-9.348 59.477-9.582 59.118Q-9.816 58.760-10.398 58.760Q-11.450 58.760-11.450 60.089Q-11.450 60.721-11.225 61.099Q-10.999 61.477-10.398 61.477M-8.131 62.191Q-8.131 61.945-7.935 61.761Q-7.738 61.576-7.482 61.497Q-7.619 61.385-7.690 61.224Q-7.762 61.063-7.762 60.882Q-7.762 60.561-7.550 60.315Q-7.885 60.017-7.885 59.607Q-7.885 59.146-7.496 58.859Q-7.106 58.572-6.628 58.572Q-6.156 58.572-5.821 58.818Q-5.647 58.664-5.436 58.582Q-5.226 58.500-4.997 58.500Q-4.833 58.500-4.712 58.607Q-4.590 58.715-4.590 58.879Q-4.590 58.975-4.662 59.047Q-4.734 59.118-4.826 59.118Q-4.925 59.118-4.995 59.045Q-5.065 58.971-5.065 58.872Q-5.065 58.818-5.052 58.787L-5.045 58.773Q-5.038 58.753-5.030 58.742Q-5.021 58.732-5.018 58.725Q-5.373 58.725-5.660 58.948Q-5.373 59.241-5.373 59.607Q-5.373 59.922-5.558 60.154Q-5.742 60.387-6.031 60.515Q-6.320 60.643-6.628 60.643Q-6.829 60.643-7.021 60.593Q-7.212 60.544-7.390 60.434Q-7.482 60.561-7.482 60.704Q-7.482 60.886-7.354 61.021Q-7.226 61.156-7.041 61.156L-6.409 61.156Q-5.961 61.156-5.592 61.227Q-5.223 61.299-4.963 61.528Q-4.703 61.757-4.703 62.191Q-4.703 62.512-4.999 62.714Q-5.294 62.916-5.698 63.005Q-6.101 63.094-6.416 63.094Q-6.733 63.094-7.137 63.005Q-7.540 62.916-7.836 62.714Q-8.131 62.512-8.131 62.191M-7.677 62.191Q-7.677 62.420-7.458 62.569Q-7.239 62.718-6.947 62.786Q-6.655 62.854-6.416 62.854Q-6.252 62.854-6.043 62.818Q-5.835 62.783-5.628 62.702Q-5.421 62.622-5.289 62.494Q-5.158 62.366-5.158 62.191Q-5.158 61.839-5.539 61.745Q-5.920 61.651-6.422 61.651L-7.041 61.651Q-7.280 61.651-7.479 61.802Q-7.677 61.952-7.677 62.191M-6.628 60.404Q-5.961 60.404-5.961 59.607Q-5.961 58.807-6.628 58.807Q-7.297 58.807-7.297 59.607Q-7.297 60.404-6.628 60.404\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(-20.365 -123.442)\">\u003Cpath d=\"M-1.062 62.793Q-0.932 62.861-0.795 62.861Q-0.624 62.861-0.474 62.772Q-0.323 62.683-0.212 62.538Q-0.101 62.393-0.023 62.225L0.241 61.658L-0.928 59.132Q-1.003 58.985-1.133 58.953Q-1.263 58.920-1.496 58.920L-1.496 58.640L0.025 58.640L0.025 58.920Q-0.323 58.920-0.323 59.067Q-0.320 59.088-0.318 59.105Q-0.316 59.122-0.316 59.132L0.541 60.991L1.314 59.320Q1.348 59.252 1.348 59.173Q1.348 59.060 1.264 58.990Q1.181 58.920 1.068 58.920L1.068 58.640L2.264 58.640L2.264 58.920Q2.045 58.920 1.873 59.024Q1.700 59.129 1.608 59.320L0.271 62.225Q0.101 62.595-0.169 62.841Q-0.440 63.087-0.795 63.087Q-1.065 63.087-1.284 62.921Q-1.503 62.755-1.503 62.492Q-1.503 62.355-1.410 62.266Q-1.318 62.178-1.178 62.178Q-1.041 62.178-0.952 62.266Q-0.863 62.355-0.863 62.492Q-0.863 62.595-0.916 62.673Q-0.969 62.752-1.062 62.793M4.934 63.408Q4.383 63.008 4.012 62.453Q3.642 61.897 3.460 61.251Q3.279 60.605 3.279 59.908Q3.279 59.395 3.380 58.900Q3.481 58.404 3.686 57.953Q3.891 57.502 4.204 57.110Q4.517 56.719 4.934 56.415Q4.944 56.411 4.951 56.410Q4.957 56.408 4.968 56.408L5.036 56.408Q5.070 56.408 5.092 56.432Q5.115 56.456 5.115 56.493Q5.115 56.538 5.087 56.555Q4.739 56.856 4.486 57.240Q4.233 57.625 4.081 58.066Q3.929 58.507 3.857 58.963Q3.785 59.419 3.785 59.908Q3.785 60.909 4.094 61.796Q4.404 62.683 5.087 63.268Q5.115 63.285 5.115 63.329Q5.115 63.367 5.092 63.391Q5.070 63.415 5.036 63.415L4.968 63.415Q4.961 63.411 4.952 63.410Q4.944 63.408 4.934 63.408M7.593 61.658L5.990 61.658L5.990 61.378Q6.215 61.378 6.364 61.344Q6.513 61.309 6.513 61.169L6.513 57.550Q6.513 57.280 6.405 57.218Q6.297 57.157 5.990 57.157L5.990 56.876L7.066 56.801L7.066 61.169Q7.066 61.306 7.217 61.342Q7.367 61.378 7.593 61.378L7.593 61.658M8.146 60.175Q8.146 59.833 8.281 59.534Q8.416 59.235 8.656 59.011Q8.895 58.787 9.213 58.662Q9.531 58.537 9.862 58.537Q10.307 58.537 10.706 58.753Q11.106 58.968 11.341 59.346Q11.575 59.723 11.575 60.175Q11.575 60.516 11.433 60.800Q11.291 61.084 11.047 61.291Q10.802 61.497 10.493 61.612Q10.184 61.726 9.862 61.726Q9.432 61.726 9.030 61.525Q8.628 61.323 8.387 60.971Q8.146 60.619 8.146 60.175M9.862 61.477Q10.464 61.477 10.688 61.099Q10.912 60.721 10.912 60.089Q10.912 59.477 10.677 59.118Q10.443 58.760 9.862 58.760Q8.810 58.760 8.810 60.089Q8.810 60.721 9.035 61.099Q9.261 61.477 9.862 61.477M13.851 61.658L12.217 61.658L12.217 61.378Q12.446 61.378 12.595 61.344Q12.744 61.309 12.744 61.169L12.744 59.320Q12.744 59.050 12.636 58.989Q12.528 58.927 12.217 58.927L12.217 58.647L13.277 58.572L13.277 59.221Q13.448 58.913 13.752 58.742Q14.056 58.572 14.401 58.572Q14.907 58.572 15.191 58.795Q15.475 59.019 15.475 59.515L15.475 61.169Q15.475 61.306 15.623 61.342Q15.772 61.378 15.997 61.378L15.997 61.658L14.367 61.658L14.367 61.378Q14.596 61.378 14.745 61.344Q14.893 61.309 14.893 61.169L14.893 59.529Q14.893 59.194 14.774 58.994Q14.654 58.794 14.340 58.794Q14.070 58.794 13.836 58.930Q13.601 59.067 13.463 59.301Q13.325 59.535 13.325 59.809L13.325 61.169Q13.325 61.306 13.475 61.342Q13.625 61.378 13.851 61.378L13.851 61.658M16.544 62.191Q16.544 61.945 16.741 61.761Q16.937 61.576 17.194 61.497Q17.057 61.385 16.985 61.224Q16.914 61.063 16.914 60.882Q16.914 60.561 17.125 60.315Q16.790 60.017 16.790 59.607Q16.790 59.146 17.180 58.859Q17.570 58.572 18.048 58.572Q18.520 58.572 18.855 58.818Q19.029 58.664 19.239 58.582Q19.450 58.500 19.679 58.500Q19.843 58.500 19.964 58.607Q20.085 58.715 20.085 58.879Q20.085 58.975 20.014 59.047Q19.942 59.118 19.850 59.118Q19.750 59.118 19.680 59.045Q19.610 58.971 19.610 58.872Q19.610 58.818 19.624 58.787L19.631 58.773Q19.638 58.753 19.646 58.742Q19.655 58.732 19.658 58.725Q19.303 58.725 19.016 58.948Q19.303 59.241 19.303 59.607Q19.303 59.922 19.118 60.154Q18.934 60.387 18.645 60.515Q18.356 60.643 18.048 60.643Q17.847 60.643 17.655 60.593Q17.464 60.544 17.286 60.434Q17.194 60.561 17.194 60.704Q17.194 60.886 17.322 61.021Q17.450 61.156 17.635 61.156L18.267 61.156Q18.715 61.156 19.084 61.227Q19.453 61.299 19.713 61.528Q19.973 61.757 19.973 62.191Q19.973 62.512 19.677 62.714Q19.381 62.916 18.978 63.005Q18.575 63.094 18.260 63.094Q17.942 63.094 17.539 63.005Q17.136 62.916 16.840 62.714Q16.544 62.512 16.544 62.191M16.999 62.191Q16.999 62.420 17.218 62.569Q17.436 62.718 17.729 62.786Q18.021 62.854 18.260 62.854Q18.424 62.854 18.633 62.818Q18.841 62.783 19.048 62.702Q19.255 62.622 19.386 62.494Q19.518 62.366 19.518 62.191Q19.518 61.839 19.137 61.745Q18.756 61.651 18.253 61.651L17.635 61.651Q17.395 61.651 17.197 61.802Q16.999 61.952 16.999 62.191M18.048 60.404Q18.715 60.404 18.715 59.607Q18.715 58.807 18.048 58.807Q17.378 58.807 17.378 59.607Q17.378 60.404 18.048 60.404M20.889 63.415L20.820 63.415Q20.786 63.415 20.764 63.389Q20.742 63.364 20.742 63.329Q20.742 63.285 20.772 63.268Q21.128 62.964 21.377 62.574Q21.627 62.184 21.779 61.752Q21.931 61.320 22.001 60.851Q22.071 60.383 22.071 59.908Q22.071 59.429 22.001 58.963Q21.931 58.496 21.777 58.061Q21.623 57.625 21.372 57.237Q21.121 56.849 20.772 56.555Q20.742 56.538 20.742 56.493Q20.742 56.459 20.764 56.434Q20.786 56.408 20.820 56.408L20.889 56.408Q20.899 56.408 20.907 56.410Q20.916 56.411 20.926 56.415Q21.470 56.815 21.842 57.368Q22.215 57.922 22.396 58.568Q22.577 59.214 22.577 59.908Q22.577 60.609 22.396 61.256Q22.215 61.904 21.841 62.458Q21.466 63.012 20.926 63.408Q20.916 63.408 20.907 63.410Q20.899 63.411 20.889 63.415\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-17.336 52.922V33.36\"\u002F>\u003Cpath stroke=\"none\" d=\"m-17.336 30.76-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-17.336 7.198v-19.563\"\u002F>\u003Cpath stroke=\"none\" d=\"m-17.336-14.965-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-17.336-32.436v-19.762\"\u002F>\u003Cpath stroke=\"none\" d=\"m-17.336-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M46.682 70.194H72.29V53.122H46.682Z\"\u002F>\u003Cg transform=\"translate(69.496 1.75)\">\u003Cpath d=\"M-15.354 61.658L-16.957 61.658L-16.957 61.378Q-16.731 61.378-16.582 61.344Q-16.434 61.309-16.434 61.169L-16.434 57.550Q-16.434 57.280-16.541 57.218Q-16.649 57.157-16.957 57.157L-16.957 56.876L-15.880 56.801L-15.880 61.169Q-15.880 61.306-15.730 61.342Q-15.579 61.378-15.354 61.378L-15.354 61.658M-14.800 60.175Q-14.800 59.833-14.665 59.534Q-14.530 59.235-14.291 59.011Q-14.051 58.787-13.733 58.662Q-13.416 58.537-13.084 58.537Q-12.640 58.537-12.240 58.753Q-11.840 58.968-11.606 59.346Q-11.372 59.723-11.372 60.175Q-11.372 60.516-11.513 60.800Q-11.655 61.084-11.900 61.291Q-12.144 61.497-12.453 61.612Q-12.763 61.726-13.084 61.726Q-13.515 61.726-13.916 61.525Q-14.318 61.323-14.559 60.971Q-14.800 60.619-14.800 60.175M-13.084 61.477Q-12.482 61.477-12.259 61.099Q-12.035 60.721-12.035 60.089Q-12.035 59.477-12.269 59.118Q-12.503 58.760-13.084 58.760Q-14.137 58.760-14.137 60.089Q-14.137 60.721-13.911 61.099Q-13.686 61.477-13.084 61.477M-9.095 61.658L-10.729 61.658L-10.729 61.378Q-10.500 61.378-10.351 61.344Q-10.203 61.309-10.203 61.169L-10.203 59.320Q-10.203 59.050-10.310 58.989Q-10.418 58.927-10.729 58.927L-10.729 58.647L-9.669 58.572L-9.669 59.221Q-9.499 58.913-9.194 58.742Q-8.890 58.572-8.545 58.572Q-8.039 58.572-7.755 58.795Q-7.472 59.019-7.472 59.515L-7.472 61.169Q-7.472 61.306-7.323 61.342Q-7.174 61.378-6.949 61.378L-6.949 61.658L-8.579 61.658L-8.579 61.378Q-8.350 61.378-8.201 61.344Q-8.053 61.309-8.053 61.169L-8.053 59.529Q-8.053 59.194-8.172 58.994Q-8.292 58.794-8.607 58.794Q-8.877 58.794-9.111 58.930Q-9.345 59.067-9.483 59.301Q-9.622 59.535-9.622 59.809L-9.622 61.169Q-9.622 61.306-9.471 61.342Q-9.321 61.378-9.095 61.378L-9.095 61.658M-6.402 62.191Q-6.402 61.945-6.205 61.761Q-6.009 61.576-5.753 61.497Q-5.889 61.385-5.961 61.224Q-6.033 61.063-6.033 60.882Q-6.033 60.561-5.821 60.315Q-6.156 60.017-6.156 59.607Q-6.156 59.146-5.766 58.859Q-5.377 58.572-4.898 58.572Q-4.426 58.572-4.091 58.818Q-3.917 58.664-3.707 58.582Q-3.497 58.500-3.268 58.500Q-3.104 58.500-2.982 58.607Q-2.861 58.715-2.861 58.879Q-2.861 58.975-2.933 59.047Q-3.004 59.118-3.097 59.118Q-3.196 59.118-3.266 59.045Q-3.336 58.971-3.336 58.872Q-3.336 58.818-3.322 58.787L-3.315 58.773Q-3.309 58.753-3.300 58.742Q-3.292 58.732-3.288 58.725Q-3.644 58.725-3.931 58.948Q-3.644 59.241-3.644 59.607Q-3.644 59.922-3.828 60.154Q-4.013 60.387-4.302 60.515Q-4.590 60.643-4.898 60.643Q-5.100 60.643-5.291 60.593Q-5.482 60.544-5.660 60.434Q-5.753 60.561-5.753 60.704Q-5.753 60.886-5.624 61.021Q-5.496 61.156-5.312 61.156L-4.679 61.156Q-4.232 61.156-3.862 61.227Q-3.493 61.299-3.233 61.528Q-2.974 61.757-2.974 62.191Q-2.974 62.512-3.269 62.714Q-3.565 62.916-3.968 63.005Q-4.372 63.094-4.686 63.094Q-5.004 63.094-5.407 63.005Q-5.811 62.916-6.106 62.714Q-6.402 62.512-6.402 62.191M-5.947 62.191Q-5.947 62.420-5.729 62.569Q-5.510 62.718-5.218 62.786Q-4.925 62.854-4.686 62.854Q-4.522 62.854-4.314 62.818Q-4.105 62.783-3.898 62.702Q-3.691 62.622-3.560 62.494Q-3.428 62.366-3.428 62.191Q-3.428 61.839-3.809 61.745Q-4.190 61.651-4.693 61.651L-5.312 61.651Q-5.551 61.651-5.749 61.802Q-5.947 61.952-5.947 62.191M-4.898 60.404Q-4.232 60.404-4.232 59.607Q-4.232 58.807-4.898 58.807Q-5.568 58.807-5.568 59.607Q-5.568 60.404-4.898 60.404\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M46.682 30.36H72.29V7.597H46.682Z\"\u002F>\u003Cg transform=\"translate(74.253 -39.554)\">\u003Cpath d=\"M-14.950 61.658L-17.037 61.658L-17.037 61.342Q-16.730 61.342-16.538 61.289Q-16.347 61.236-16.347 61.047L-16.347 56.332Q-16.347 56.090-16.418 55.982Q-16.488 55.875-16.622 55.851Q-16.756 55.826-17.037 55.826L-17.037 55.510L-15.670 55.413L-15.670 58.463Q-15.473 58.107-15.119 57.894Q-14.765 57.681-14.365 57.681Q-13.086 57.681-13.086 58.894L-13.086 61.047Q-13.086 61.236-12.895 61.289Q-12.704 61.342-12.397 61.342L-12.397 61.658L-14.484 61.658L-14.484 61.342Q-14.172 61.342-13.981 61.289Q-13.790 61.236-13.790 61.047L-13.790 58.929Q-13.790 58.670-13.834 58.448Q-13.878 58.226-14.023 58.083Q-14.168 57.940-14.427 57.940Q-14.770 57.940-15.051 58.129Q-15.332 58.318-15.488 58.630Q-15.644 58.942-15.644 59.289L-15.644 61.047Q-15.644 61.236-15.451 61.289Q-15.257 61.342-14.950 61.342\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M42.395-15.165h34.182v-17.071H42.395Z\"\u002F>\u003Cg transform=\"translate(63.065 -82.928)\">\u003Cpath d=\"M-17.022 61.651L-17.022 60.588Q-17.022 60.564-16.994 60.537Q-16.967 60.510-16.943 60.510L-16.834 60.510Q-16.769 60.510-16.755 60.568Q-16.659 61.002-16.413 61.253Q-16.167 61.504-15.753 61.504Q-15.412 61.504-15.159 61.371Q-14.906 61.238-14.906 60.930Q-14.906 60.773-15 60.658Q-15.094 60.544-15.232 60.475Q-15.371 60.407-15.538 60.369L-16.119 60.270Q-16.475 60.202-16.748 59.981Q-17.022 59.761-17.022 59.419Q-17.022 59.170-16.910 58.995Q-16.799 58.821-16.613 58.722Q-16.427 58.623-16.211 58.580Q-15.996 58.537-15.753 58.537Q-15.340 58.537-15.060 58.719L-14.844 58.544Q-14.834 58.541-14.827 58.539Q-14.820 58.537-14.810 58.537L-14.759 58.537Q-14.732 58.537-14.708 58.561Q-14.684 58.585-14.684 58.613L-14.684 59.460Q-14.684 59.481-14.708 59.508Q-14.732 59.535-14.759 59.535L-14.872 59.535Q-14.899 59.535-14.925 59.510Q-14.950 59.484-14.950 59.460Q-14.950 59.224-15.056 59.060Q-15.162 58.896-15.345 58.814Q-15.528 58.732-15.760 58.732Q-16.088 58.732-16.345 58.835Q-16.601 58.937-16.601 59.214Q-16.601 59.409-16.418 59.518Q-16.235 59.628-16.006 59.669L-15.432 59.775Q-15.186 59.823-14.972 59.951Q-14.759 60.079-14.622 60.282Q-14.485 60.486-14.485 60.735Q-14.485 61.248-14.851 61.487Q-15.217 61.726-15.753 61.726Q-16.249 61.726-16.581 61.432L-16.847 61.706Q-16.868 61.726-16.895 61.726L-16.943 61.726Q-16.967 61.726-16.994 61.699Q-17.022 61.672-17.022 61.651M-13.898 60.175Q-13.898 59.833-13.763 59.534Q-13.628 59.235-13.388 59.011Q-13.149 58.787-12.831 58.662Q-12.513 58.537-12.182 58.537Q-11.737 58.537-11.337 58.753Q-10.938 58.968-10.703 59.346Q-10.469 59.723-10.469 60.175Q-10.469 60.516-10.611 60.800Q-10.753 61.084-10.997 61.291Q-11.242 61.497-11.551 61.612Q-11.860 61.726-12.182 61.726Q-12.612 61.726-13.014 61.525Q-13.416 61.323-13.657 60.971Q-13.898 60.619-13.898 60.175M-12.182 61.477Q-11.580 61.477-11.356 61.099Q-11.132 60.721-11.132 60.089Q-11.132 59.477-11.367 59.118Q-11.601 58.760-12.182 58.760Q-13.234 58.760-13.234 60.089Q-13.234 60.721-13.009 61.099Q-12.783 61.477-12.182 61.477M-8.077 61.658L-9.810 61.658L-9.810 61.378Q-9.584 61.378-9.435 61.344Q-9.287 61.309-9.287 61.169L-9.287 58.920L-9.875 58.920L-9.875 58.640L-9.287 58.640L-9.287 57.823Q-9.287 57.505-9.109 57.257Q-8.931 57.010-8.641 56.869Q-8.350 56.729-8.039 56.729Q-7.783 56.729-7.579 56.871Q-7.376 57.013-7.376 57.256Q-7.376 57.392-7.475 57.491Q-7.574 57.591-7.711 57.591Q-7.848 57.591-7.947 57.491Q-8.046 57.392-8.046 57.256Q-8.046 57.075-7.906 56.982Q-7.984 56.955-8.084 56.955Q-8.292 56.955-8.446 57.088Q-8.600 57.221-8.680 57.425Q-8.760 57.628-8.760 57.837L-8.760 58.640L-7.872 58.640L-7.872 58.920L-8.733 58.920L-8.733 61.169Q-8.733 61.378-8.077 61.378L-8.077 61.658M-6.870 60.817L-6.870 58.920L-7.509 58.920L-7.509 58.698Q-7.191 58.698-6.974 58.488Q-6.757 58.278-6.657 57.968Q-6.556 57.659-6.556 57.351L-6.289 57.351L-6.289 58.640L-5.212 58.640L-5.212 58.920L-6.289 58.920L-6.289 60.804Q-6.289 61.080-6.185 61.279Q-6.081 61.477-5.821 61.477Q-5.664 61.477-5.558 61.373Q-5.452 61.268-5.402 61.115Q-5.353 60.961-5.353 60.804L-5.353 60.390L-5.086 60.390L-5.086 60.817Q-5.086 61.043-5.185 61.253Q-5.284 61.463-5.469 61.595Q-5.653 61.726-5.882 61.726Q-6.320 61.726-6.595 61.489Q-6.870 61.251-6.870 60.817M-2.594 61.658L-4.228 61.658L-4.228 61.378Q-3.999 61.378-3.850 61.344Q-3.702 61.309-3.702 61.169L-3.702 59.320Q-3.702 59.050-3.809 58.989Q-3.917 58.927-4.228 58.927L-4.228 58.647L-3.169 58.572L-3.169 59.221Q-2.998 58.913-2.693 58.742Q-2.389 58.572-2.044 58.572Q-1.644 58.572-1.367 58.712Q-1.090 58.852-1.005 59.200Q-0.837 58.907-0.538 58.739Q-0.239 58.572 0.106 58.572Q0.612 58.572 0.895 58.795Q1.179 59.019 1.179 59.515L1.179 61.169Q1.179 61.306 1.328 61.342Q1.477 61.378 1.702 61.378L1.702 61.658L0.072 61.658L0.072 61.378Q0.297 61.378 0.448 61.342Q0.598 61.306 0.598 61.169L0.598 59.529Q0.598 59.194 0.478 58.994Q0.359 58.794 0.044 58.794Q-0.226 58.794-0.460 58.930Q-0.694 59.067-0.832 59.301Q-0.971 59.535-0.971 59.809L-0.971 61.169Q-0.971 61.306-0.822 61.342Q-0.673 61.378-0.448 61.378L-0.448 61.658L-2.078 61.658L-2.078 61.378Q-1.849 61.378-1.701 61.344Q-1.552 61.309-1.552 61.169L-1.552 59.529Q-1.552 59.194-1.671 58.994Q-1.791 58.794-2.106 58.794Q-2.376 58.794-2.610 58.930Q-2.844 59.067-2.982 59.301Q-3.121 59.535-3.121 59.809L-3.121 61.169Q-3.121 61.306-2.970 61.342Q-2.820 61.378-2.594 61.378L-2.594 61.658M2.348 60.930Q2.348 60.598 2.572 60.371Q2.796 60.144 3.139 60.016Q3.483 59.887 3.855 59.835Q4.228 59.782 4.532 59.782L4.532 59.529Q4.532 59.324 4.424 59.144Q4.317 58.965 4.136 58.862Q3.955 58.760 3.746 58.760Q3.339 58.760 3.103 58.852Q3.192 58.889 3.238 58.973Q3.285 59.057 3.285 59.159Q3.285 59.255 3.238 59.334Q3.192 59.412 3.112 59.457Q3.032 59.501 2.943 59.501Q2.792 59.501 2.692 59.404Q2.591 59.306 2.591 59.159Q2.591 58.537 3.746 58.537Q3.958 58.537 4.207 58.601Q4.457 58.664 4.659 58.783Q4.860 58.903 4.987 59.088Q5.113 59.272 5.113 59.515L5.113 61.091Q5.113 61.207 5.175 61.303Q5.236 61.398 5.349 61.398Q5.458 61.398 5.523 61.304Q5.588 61.210 5.588 61.091L5.588 60.643L5.855 60.643L5.855 61.091Q5.855 61.361 5.628 61.526Q5.400 61.692 5.120 61.692Q4.912 61.692 4.775 61.538Q4.638 61.385 4.614 61.169Q4.467 61.436 4.185 61.581Q3.903 61.726 3.579 61.726Q3.302 61.726 3.018 61.651Q2.734 61.576 2.541 61.397Q2.348 61.217 2.348 60.930M2.963 60.930Q2.963 61.104 3.064 61.234Q3.165 61.364 3.320 61.434Q3.476 61.504 3.640 61.504Q3.859 61.504 4.067 61.407Q4.276 61.309 4.404 61.128Q4.532 60.947 4.532 60.721L4.532 59.993Q4.207 59.993 3.842 60.084Q3.476 60.175 3.220 60.387Q2.963 60.598 2.963 60.930M7.455 61.658L6.132 61.658L6.132 61.378Q6.692 61.378 7.072 60.978L7.786 60.181L6.873 59.132Q6.737 58.985 6.588 58.953Q6.439 58.920 6.173 58.920L6.173 58.640L7.673 58.640L7.673 58.920Q7.482 58.920 7.482 59.054Q7.482 59.084 7.513 59.132L8.107 59.816L8.548 59.320Q8.661 59.190 8.661 59.074Q8.661 59.012 8.623 58.966Q8.586 58.920 8.528 58.920L8.528 58.640L9.844 58.640L9.844 58.920Q9.283 58.920 8.904 59.320L8.282 60.021L9.276 61.169Q9.375 61.268 9.476 61.313Q9.577 61.357 9.688 61.367Q9.799 61.378 9.977 61.378L9.977 61.658L8.483 61.658L8.483 61.378Q8.548 61.378 8.608 61.344Q8.668 61.309 8.668 61.244Q8.668 61.197 8.637 61.169L7.960 60.383L7.427 60.978Q7.314 61.108 7.314 61.224Q7.314 61.289 7.355 61.333Q7.396 61.378 7.455 61.378\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M36.705-54.998h45.562V-72.07H36.705Z\"\u002F>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(57.374 -123.442)\">\u003Cpath d=\"M-15.138 60.404L-17.196 60.404L-17.196 59.901L-15.138 59.901L-15.138 60.404M-12.667 61.658L-14.270 61.658L-14.270 61.378Q-14.044 61.378-13.896 61.344Q-13.747 61.309-13.747 61.169L-13.747 57.550Q-13.747 57.280-13.855 57.218Q-13.962 57.157-14.270 57.157L-14.270 56.876L-13.193 56.801L-13.193 61.169Q-13.193 61.306-13.043 61.342Q-12.893 61.378-12.667 61.378L-12.667 61.658M-12.113 60.175Q-12.113 59.833-11.978 59.534Q-11.843 59.235-11.604 59.011Q-11.365 58.787-11.047 58.662Q-10.729 58.537-10.398 58.537Q-9.953 58.537-9.553 58.753Q-9.153 58.968-8.919 59.346Q-8.685 59.723-8.685 60.175Q-8.685 60.516-8.827 60.800Q-8.969 61.084-9.213 61.291Q-9.458 61.497-9.767 61.612Q-10.076 61.726-10.398 61.726Q-10.828 61.726-11.230 61.525Q-11.631 61.323-11.872 60.971Q-12.113 60.619-12.113 60.175M-10.398 61.477Q-9.796 61.477-9.572 61.099Q-9.348 60.721-9.348 60.089Q-9.348 59.477-9.582 59.118Q-9.816 58.760-10.398 58.760Q-11.450 58.760-11.450 60.089Q-11.450 60.721-11.225 61.099Q-10.999 61.477-10.398 61.477M-8.131 62.191Q-8.131 61.945-7.935 61.761Q-7.738 61.576-7.482 61.497Q-7.619 61.385-7.690 61.224Q-7.762 61.063-7.762 60.882Q-7.762 60.561-7.550 60.315Q-7.885 60.017-7.885 59.607Q-7.885 59.146-7.496 58.859Q-7.106 58.572-6.628 58.572Q-6.156 58.572-5.821 58.818Q-5.647 58.664-5.436 58.582Q-5.226 58.500-4.997 58.500Q-4.833 58.500-4.712 58.607Q-4.590 58.715-4.590 58.879Q-4.590 58.975-4.662 59.047Q-4.734 59.118-4.826 59.118Q-4.925 59.118-4.995 59.045Q-5.065 58.971-5.065 58.872Q-5.065 58.818-5.052 58.787L-5.045 58.773Q-5.038 58.753-5.030 58.742Q-5.021 58.732-5.018 58.725Q-5.373 58.725-5.660 58.948Q-5.373 59.241-5.373 59.607Q-5.373 59.922-5.558 60.154Q-5.742 60.387-6.031 60.515Q-6.320 60.643-6.628 60.643Q-6.829 60.643-7.021 60.593Q-7.212 60.544-7.390 60.434Q-7.482 60.561-7.482 60.704Q-7.482 60.886-7.354 61.021Q-7.226 61.156-7.041 61.156L-6.409 61.156Q-5.961 61.156-5.592 61.227Q-5.223 61.299-4.963 61.528Q-4.703 61.757-4.703 62.191Q-4.703 62.512-4.999 62.714Q-5.294 62.916-5.698 63.005Q-6.101 63.094-6.416 63.094Q-6.733 63.094-7.137 63.005Q-7.540 62.916-7.836 62.714Q-8.131 62.512-8.131 62.191M-7.677 62.191Q-7.677 62.420-7.458 62.569Q-7.239 62.718-6.947 62.786Q-6.655 62.854-6.416 62.854Q-6.252 62.854-6.043 62.818Q-5.835 62.783-5.628 62.702Q-5.421 62.622-5.289 62.494Q-5.158 62.366-5.158 62.191Q-5.158 61.839-5.539 61.745Q-5.920 61.651-6.422 61.651L-7.041 61.651Q-7.280 61.651-7.479 61.802Q-7.677 61.952-7.677 62.191M-6.628 60.404Q-5.961 60.404-5.961 59.607Q-5.961 58.807-6.628 58.807Q-7.297 58.807-7.297 59.607Q-7.297 60.404-6.628 60.404\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(57.374 -123.442)\">\u003Cpath d=\"M-1.062 62.793Q-0.932 62.861-0.795 62.861Q-0.624 62.861-0.474 62.772Q-0.323 62.683-0.212 62.538Q-0.101 62.393-0.023 62.225L0.241 61.658L-0.928 59.132Q-1.003 58.985-1.133 58.953Q-1.263 58.920-1.496 58.920L-1.496 58.640L0.025 58.640L0.025 58.920Q-0.323 58.920-0.323 59.067Q-0.320 59.088-0.318 59.105Q-0.316 59.122-0.316 59.132L0.541 60.991L1.314 59.320Q1.348 59.252 1.348 59.173Q1.348 59.060 1.264 58.990Q1.181 58.920 1.068 58.920L1.068 58.640L2.264 58.640L2.264 58.920Q2.045 58.920 1.873 59.024Q1.700 59.129 1.608 59.320L0.271 62.225Q0.101 62.595-0.169 62.841Q-0.440 63.087-0.795 63.087Q-1.065 63.087-1.284 62.921Q-1.503 62.755-1.503 62.492Q-1.503 62.355-1.410 62.266Q-1.318 62.178-1.178 62.178Q-1.041 62.178-0.952 62.266Q-0.863 62.355-0.863 62.492Q-0.863 62.595-0.916 62.673Q-0.969 62.752-1.062 62.793M4.934 63.408Q4.383 63.008 4.012 62.453Q3.642 61.897 3.460 61.251Q3.279 60.605 3.279 59.908Q3.279 59.395 3.380 58.900Q3.481 58.404 3.686 57.953Q3.891 57.502 4.204 57.110Q4.517 56.719 4.934 56.415Q4.944 56.411 4.951 56.410Q4.957 56.408 4.968 56.408L5.036 56.408Q5.070 56.408 5.092 56.432Q5.115 56.456 5.115 56.493Q5.115 56.538 5.087 56.555Q4.739 56.856 4.486 57.240Q4.233 57.625 4.081 58.066Q3.929 58.507 3.857 58.963Q3.785 59.419 3.785 59.908Q3.785 60.909 4.094 61.796Q4.404 62.683 5.087 63.268Q5.115 63.285 5.115 63.329Q5.115 63.367 5.092 63.391Q5.070 63.415 5.036 63.415L4.968 63.415Q4.961 63.411 4.952 63.410Q4.944 63.408 4.934 63.408M5.983 60.930Q5.983 60.598 6.207 60.371Q6.431 60.144 6.774 60.016Q7.118 59.887 7.490 59.835Q7.863 59.782 8.167 59.782L8.167 59.529Q8.167 59.324 8.059 59.144Q7.952 58.965 7.770 58.862Q7.589 58.760 7.381 58.760Q6.974 58.760 6.738 58.852Q6.827 58.889 6.873 58.973Q6.919 59.057 6.919 59.159Q6.919 59.255 6.873 59.334Q6.827 59.412 6.747 59.457Q6.666 59.501 6.578 59.501Q6.427 59.501 6.326 59.404Q6.226 59.306 6.226 59.159Q6.226 58.537 7.381 58.537Q7.593 58.537 7.842 58.601Q8.092 58.664 8.293 58.783Q8.495 58.903 8.622 59.088Q8.748 59.272 8.748 59.515L8.748 61.091Q8.748 61.207 8.810 61.303Q8.871 61.398 8.984 61.398Q9.093 61.398 9.158 61.304Q9.223 61.210 9.223 61.091L9.223 60.643L9.490 60.643L9.490 61.091Q9.490 61.361 9.262 61.526Q9.035 61.692 8.755 61.692Q8.546 61.692 8.410 61.538Q8.273 61.385 8.249 61.169Q8.102 61.436 7.820 61.581Q7.538 61.726 7.213 61.726Q6.936 61.726 6.653 61.651Q6.369 61.576 6.176 61.397Q5.983 61.217 5.983 60.930M6.598 60.930Q6.598 61.104 6.699 61.234Q6.800 61.364 6.955 61.434Q7.111 61.504 7.275 61.504Q7.494 61.504 7.702 61.407Q7.911 61.309 8.039 61.128Q8.167 60.947 8.167 60.721L8.167 59.993Q7.842 59.993 7.476 60.084Q7.111 60.175 6.854 60.387Q6.598 60.598 6.598 60.930M11.588 61.658L9.955 61.658L9.955 61.378Q10.184 61.378 10.332 61.344Q10.481 61.309 10.481 61.169L10.481 59.320Q10.481 59.050 10.373 58.989Q10.266 58.927 9.955 58.927L9.955 58.647L11.014 58.572L11.014 59.221Q11.185 58.913 11.489 58.742Q11.793 58.572 12.139 58.572Q12.644 58.572 12.928 58.795Q13.212 59.019 13.212 59.515L13.212 61.169Q13.212 61.306 13.361 61.342Q13.509 61.378 13.735 61.378L13.735 61.658L12.104 61.658L12.104 61.378Q12.333 61.378 12.482 61.344Q12.631 61.309 12.631 61.169L12.631 59.529Q12.631 59.194 12.511 58.994Q12.392 58.794 12.077 58.794Q11.807 58.794 11.573 58.930Q11.339 59.067 11.200 59.301Q11.062 59.535 11.062 59.809L11.062 61.169Q11.062 61.306 11.212 61.342Q11.363 61.378 11.588 61.378L11.588 61.658M14.323 60.147Q14.323 59.809 14.463 59.518Q14.603 59.228 14.847 59.014Q15.092 58.801 15.396 58.686Q15.700 58.572 16.025 58.572Q16.295 58.572 16.558 58.671Q16.821 58.770 17.013 58.948L17.013 57.550Q17.013 57.280 16.905 57.218Q16.797 57.157 16.486 57.157L16.486 56.876L17.563 56.801L17.563 60.985Q17.563 61.173 17.618 61.256Q17.672 61.340 17.773 61.359Q17.874 61.378 18.089 61.378L18.089 61.658L16.982 61.726L16.982 61.309Q16.565 61.726 15.939 61.726Q15.509 61.726 15.136 61.514Q14.764 61.303 14.543 60.942Q14.323 60.581 14.323 60.147M15.997 61.504Q16.206 61.504 16.392 61.432Q16.579 61.361 16.732 61.224Q16.886 61.087 16.982 60.909L16.982 59.300Q16.896 59.153 16.751 59.033Q16.606 58.913 16.437 58.854Q16.268 58.794 16.086 58.794Q15.526 58.794 15.258 59.183Q14.989 59.573 14.989 60.154Q14.989 60.725 15.223 61.115Q15.457 61.504 15.997 61.504M19.060 63.415L18.992 63.415Q18.957 63.415 18.935 63.389Q18.913 63.364 18.913 63.329Q18.913 63.285 18.944 63.268Q19.299 62.964 19.549 62.574Q19.798 62.184 19.950 61.752Q20.102 61.320 20.173 60.851Q20.243 60.383 20.243 59.908Q20.243 59.429 20.173 58.963Q20.102 58.496 19.949 58.061Q19.795 57.625 19.544 57.237Q19.292 56.849 18.944 56.555Q18.913 56.538 18.913 56.493Q18.913 56.459 18.935 56.434Q18.957 56.408 18.992 56.408L19.060 56.408Q19.070 56.408 19.079 56.410Q19.087 56.411 19.098 56.415Q19.641 56.815 20.014 57.368Q20.386 57.922 20.567 58.568Q20.748 59.214 20.748 59.908Q20.748 60.609 20.567 61.256Q20.386 61.904 20.012 62.458Q19.638 63.012 19.098 63.408Q19.087 63.408 19.079 63.410Q19.070 63.411 19.060 63.415\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M59.486 52.922V33.36\"\u002F>\u003Cpath stroke=\"none\" d=\"m59.486 30.76-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M59.486 7.198v-19.563\"\u002F>\u003Cpath stroke=\"none\" d=\"m59.486-14.965-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M59.486-32.436v-19.762\"\u002F>\u003Cpath stroke=\"none\" d=\"m59.486-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M123.505 70.194h25.607V53.122h-25.607Z\"\u002F>\u003Cg transform=\"translate(147.235 2.43)\">\u003Cpath d=\"M-16.963 60.930Q-16.963 60.598-16.740 60.371Q-16.516 60.144-16.172 60.016Q-15.829 59.887-15.456 59.835Q-15.084 59.782-14.779 59.782L-14.779 59.529Q-14.779 59.324-14.887 59.144Q-14.995 58.965-15.176 58.862Q-15.357 58.760-15.565 58.760Q-15.972 58.760-16.208 58.852Q-16.119 58.889-16.073 58.973Q-16.027 59.057-16.027 59.159Q-16.027 59.255-16.073 59.334Q-16.119 59.412-16.200 59.457Q-16.280 59.501-16.369 59.501Q-16.519 59.501-16.620 59.404Q-16.721 59.306-16.721 59.159Q-16.721 58.537-15.565 58.537Q-15.354 58.537-15.104 58.601Q-14.855 58.664-14.653 58.783Q-14.451 58.903-14.325 59.088Q-14.198 59.272-14.198 59.515L-14.198 61.091Q-14.198 61.207-14.137 61.303Q-14.075 61.398-13.962 61.398Q-13.853 61.398-13.788 61.304Q-13.723 61.210-13.723 61.091L-13.723 60.643L-13.457 60.643L-13.457 61.091Q-13.457 61.361-13.684 61.526Q-13.911 61.692-14.191 61.692Q-14.400 61.692-14.537 61.538Q-14.673 61.385-14.697 61.169Q-14.844 61.436-15.126 61.581Q-15.408 61.726-15.733 61.726Q-16.010 61.726-16.294 61.651Q-16.577 61.576-16.770 61.397Q-16.963 61.217-16.963 60.930M-16.348 60.930Q-16.348 61.104-16.247 61.234Q-16.147 61.364-15.991 61.434Q-15.836 61.504-15.671 61.504Q-15.453 61.504-15.244 61.407Q-15.036 61.309-14.908 61.128Q-14.779 60.947-14.779 60.721L-14.779 59.993Q-15.104 59.993-15.470 60.084Q-15.836 60.175-16.092 60.387Q-16.348 60.598-16.348 60.930M-11.358 61.658L-12.992 61.658L-12.992 61.378Q-12.763 61.378-12.614 61.344Q-12.465 61.309-12.465 61.169L-12.465 59.320Q-12.465 59.050-12.573 58.989Q-12.681 58.927-12.992 58.927L-12.992 58.647L-11.932 58.572L-11.932 59.221Q-11.761 58.913-11.457 58.742Q-11.153 58.572-10.808 58.572Q-10.302 58.572-10.018 58.795Q-9.734 59.019-9.734 59.515L-9.734 61.169Q-9.734 61.306-9.586 61.342Q-9.437 61.378-9.211 61.378L-9.211 61.658L-10.842 61.658L-10.842 61.378Q-10.613 61.378-10.464 61.344Q-10.315 61.309-10.315 61.169L-10.315 59.529Q-10.315 59.194-10.435 58.994Q-10.555 58.794-10.869 58.794Q-11.139 58.794-11.373 58.930Q-11.607 59.067-11.746 59.301Q-11.884 59.535-11.884 59.809L-11.884 61.169Q-11.884 61.306-11.734 61.342Q-11.584 61.378-11.358 61.378L-11.358 61.658M-8.624 60.147Q-8.624 59.809-8.483 59.518Q-8.343 59.228-8.099 59.014Q-7.855 58.801-7.550 58.686Q-7.246 58.572-6.921 58.572Q-6.651 58.572-6.388 58.671Q-6.125 58.770-5.934 58.948L-5.934 57.550Q-5.934 57.280-6.041 57.218Q-6.149 57.157-6.460 57.157L-6.460 56.876L-5.383 56.801L-5.383 60.985Q-5.383 61.173-5.329 61.256Q-5.274 61.340-5.173 61.359Q-5.072 61.378-4.857 61.378L-4.857 61.658L-5.964 61.726L-5.964 61.309Q-6.381 61.726-7.007 61.726Q-7.438 61.726-7.810 61.514Q-8.183 61.303-8.403 60.942Q-8.624 60.581-8.624 60.147M-6.949 61.504Q-6.740 61.504-6.554 61.432Q-6.368 61.361-6.214 61.224Q-6.060 61.087-5.964 60.909L-5.964 59.300Q-6.050 59.153-6.195 59.033Q-6.340 58.913-6.510 58.854Q-6.679 58.794-6.860 58.794Q-7.420 58.794-7.689 59.183Q-7.957 59.573-7.957 60.154Q-7.957 60.725-7.723 61.115Q-7.489 61.504-6.949 61.504\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M123.505 30.36h25.607V7.597h-25.607Z\"\u002F>\u003Cg transform=\"translate(151.075 -39.554)\">\u003Cpath d=\"M-14.950 61.658L-17.037 61.658L-17.037 61.342Q-16.730 61.342-16.538 61.289Q-16.347 61.236-16.347 61.047L-16.347 56.332Q-16.347 56.090-16.418 55.982Q-16.488 55.875-16.622 55.851Q-16.756 55.826-17.037 55.826L-17.037 55.510L-15.670 55.413L-15.670 58.463Q-15.473 58.107-15.119 57.894Q-14.765 57.681-14.365 57.681Q-13.086 57.681-13.086 58.894L-13.086 61.047Q-13.086 61.236-12.895 61.289Q-12.704 61.342-12.397 61.342L-12.397 61.658L-14.484 61.658L-14.484 61.342Q-14.172 61.342-13.981 61.289Q-13.790 61.236-13.790 61.047L-13.790 58.929Q-13.790 58.670-13.834 58.448Q-13.878 58.226-14.023 58.083Q-14.168 57.940-14.427 57.940Q-14.770 57.940-15.051 58.129Q-15.332 58.318-15.488 58.630Q-15.644 58.942-15.644 59.289L-15.644 61.047Q-15.644 61.236-15.451 61.289Q-15.257 61.342-14.950 61.342\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M119.218-15.165h34.181v-17.071h-34.181Z\"\u002F>\u003Cg transform=\"translate(139.887 -82.928)\">\u003Cpath d=\"M-17.022 61.651L-17.022 60.588Q-17.022 60.564-16.994 60.537Q-16.967 60.510-16.943 60.510L-16.834 60.510Q-16.769 60.510-16.755 60.568Q-16.659 61.002-16.413 61.253Q-16.167 61.504-15.753 61.504Q-15.412 61.504-15.159 61.371Q-14.906 61.238-14.906 60.930Q-14.906 60.773-15 60.658Q-15.094 60.544-15.232 60.475Q-15.371 60.407-15.538 60.369L-16.119 60.270Q-16.475 60.202-16.748 59.981Q-17.022 59.761-17.022 59.419Q-17.022 59.170-16.910 58.995Q-16.799 58.821-16.613 58.722Q-16.427 58.623-16.211 58.580Q-15.996 58.537-15.753 58.537Q-15.340 58.537-15.060 58.719L-14.844 58.544Q-14.834 58.541-14.827 58.539Q-14.820 58.537-14.810 58.537L-14.759 58.537Q-14.732 58.537-14.708 58.561Q-14.684 58.585-14.684 58.613L-14.684 59.460Q-14.684 59.481-14.708 59.508Q-14.732 59.535-14.759 59.535L-14.872 59.535Q-14.899 59.535-14.925 59.510Q-14.950 59.484-14.950 59.460Q-14.950 59.224-15.056 59.060Q-15.162 58.896-15.345 58.814Q-15.528 58.732-15.760 58.732Q-16.088 58.732-16.345 58.835Q-16.601 58.937-16.601 59.214Q-16.601 59.409-16.418 59.518Q-16.235 59.628-16.006 59.669L-15.432 59.775Q-15.186 59.823-14.972 59.951Q-14.759 60.079-14.622 60.282Q-14.485 60.486-14.485 60.735Q-14.485 61.248-14.851 61.487Q-15.217 61.726-15.753 61.726Q-16.249 61.726-16.581 61.432L-16.847 61.706Q-16.868 61.726-16.895 61.726L-16.943 61.726Q-16.967 61.726-16.994 61.699Q-17.022 61.672-17.022 61.651M-13.898 60.175Q-13.898 59.833-13.763 59.534Q-13.628 59.235-13.388 59.011Q-13.149 58.787-12.831 58.662Q-12.513 58.537-12.182 58.537Q-11.737 58.537-11.337 58.753Q-10.938 58.968-10.703 59.346Q-10.469 59.723-10.469 60.175Q-10.469 60.516-10.611 60.800Q-10.753 61.084-10.997 61.291Q-11.242 61.497-11.551 61.612Q-11.860 61.726-12.182 61.726Q-12.612 61.726-13.014 61.525Q-13.416 61.323-13.657 60.971Q-13.898 60.619-13.898 60.175M-12.182 61.477Q-11.580 61.477-11.356 61.099Q-11.132 60.721-11.132 60.089Q-11.132 59.477-11.367 59.118Q-11.601 58.760-12.182 58.760Q-13.234 58.760-13.234 60.089Q-13.234 60.721-13.009 61.099Q-12.783 61.477-12.182 61.477M-8.077 61.658L-9.810 61.658L-9.810 61.378Q-9.584 61.378-9.435 61.344Q-9.287 61.309-9.287 61.169L-9.287 58.920L-9.875 58.920L-9.875 58.640L-9.287 58.640L-9.287 57.823Q-9.287 57.505-9.109 57.257Q-8.931 57.010-8.641 56.869Q-8.350 56.729-8.039 56.729Q-7.783 56.729-7.579 56.871Q-7.376 57.013-7.376 57.256Q-7.376 57.392-7.475 57.491Q-7.574 57.591-7.711 57.591Q-7.848 57.591-7.947 57.491Q-8.046 57.392-8.046 57.256Q-8.046 57.075-7.906 56.982Q-7.984 56.955-8.084 56.955Q-8.292 56.955-8.446 57.088Q-8.600 57.221-8.680 57.425Q-8.760 57.628-8.760 57.837L-8.760 58.640L-7.872 58.640L-7.872 58.920L-8.733 58.920L-8.733 61.169Q-8.733 61.378-8.077 61.378L-8.077 61.658M-6.870 60.817L-6.870 58.920L-7.509 58.920L-7.509 58.698Q-7.191 58.698-6.974 58.488Q-6.757 58.278-6.657 57.968Q-6.556 57.659-6.556 57.351L-6.289 57.351L-6.289 58.640L-5.212 58.640L-5.212 58.920L-6.289 58.920L-6.289 60.804Q-6.289 61.080-6.185 61.279Q-6.081 61.477-5.821 61.477Q-5.664 61.477-5.558 61.373Q-5.452 61.268-5.402 61.115Q-5.353 60.961-5.353 60.804L-5.353 60.390L-5.086 60.390L-5.086 60.817Q-5.086 61.043-5.185 61.253Q-5.284 61.463-5.469 61.595Q-5.653 61.726-5.882 61.726Q-6.320 61.726-6.595 61.489Q-6.870 61.251-6.870 60.817M-2.594 61.658L-4.228 61.658L-4.228 61.378Q-3.999 61.378-3.850 61.344Q-3.702 61.309-3.702 61.169L-3.702 59.320Q-3.702 59.050-3.809 58.989Q-3.917 58.927-4.228 58.927L-4.228 58.647L-3.169 58.572L-3.169 59.221Q-2.998 58.913-2.693 58.742Q-2.389 58.572-2.044 58.572Q-1.644 58.572-1.367 58.712Q-1.090 58.852-1.005 59.200Q-0.837 58.907-0.538 58.739Q-0.239 58.572 0.106 58.572Q0.612 58.572 0.895 58.795Q1.179 59.019 1.179 59.515L1.179 61.169Q1.179 61.306 1.328 61.342Q1.477 61.378 1.702 61.378L1.702 61.658L0.072 61.658L0.072 61.378Q0.297 61.378 0.448 61.342Q0.598 61.306 0.598 61.169L0.598 59.529Q0.598 59.194 0.478 58.994Q0.359 58.794 0.044 58.794Q-0.226 58.794-0.460 58.930Q-0.694 59.067-0.832 59.301Q-0.971 59.535-0.971 59.809L-0.971 61.169Q-0.971 61.306-0.822 61.342Q-0.673 61.378-0.448 61.378L-0.448 61.658L-2.078 61.658L-2.078 61.378Q-1.849 61.378-1.701 61.344Q-1.552 61.309-1.552 61.169L-1.552 59.529Q-1.552 59.194-1.671 58.994Q-1.791 58.794-2.106 58.794Q-2.376 58.794-2.610 58.930Q-2.844 59.067-2.982 59.301Q-3.121 59.535-3.121 59.809L-3.121 61.169Q-3.121 61.306-2.970 61.342Q-2.820 61.378-2.594 61.378L-2.594 61.658M2.348 60.930Q2.348 60.598 2.572 60.371Q2.796 60.144 3.139 60.016Q3.483 59.887 3.855 59.835Q4.228 59.782 4.532 59.782L4.532 59.529Q4.532 59.324 4.424 59.144Q4.317 58.965 4.136 58.862Q3.955 58.760 3.746 58.760Q3.339 58.760 3.103 58.852Q3.192 58.889 3.238 58.973Q3.285 59.057 3.285 59.159Q3.285 59.255 3.238 59.334Q3.192 59.412 3.112 59.457Q3.032 59.501 2.943 59.501Q2.792 59.501 2.692 59.404Q2.591 59.306 2.591 59.159Q2.591 58.537 3.746 58.537Q3.958 58.537 4.207 58.601Q4.457 58.664 4.659 58.783Q4.860 58.903 4.987 59.088Q5.113 59.272 5.113 59.515L5.113 61.091Q5.113 61.207 5.175 61.303Q5.236 61.398 5.349 61.398Q5.458 61.398 5.523 61.304Q5.588 61.210 5.588 61.091L5.588 60.643L5.855 60.643L5.855 61.091Q5.855 61.361 5.628 61.526Q5.400 61.692 5.120 61.692Q4.912 61.692 4.775 61.538Q4.638 61.385 4.614 61.169Q4.467 61.436 4.185 61.581Q3.903 61.726 3.579 61.726Q3.302 61.726 3.018 61.651Q2.734 61.576 2.541 61.397Q2.348 61.217 2.348 60.930M2.963 60.930Q2.963 61.104 3.064 61.234Q3.165 61.364 3.320 61.434Q3.476 61.504 3.640 61.504Q3.859 61.504 4.067 61.407Q4.276 61.309 4.404 61.128Q4.532 60.947 4.532 60.721L4.532 59.993Q4.207 59.993 3.842 60.084Q3.476 60.175 3.220 60.387Q2.963 60.598 2.963 60.930M7.455 61.658L6.132 61.658L6.132 61.378Q6.692 61.378 7.072 60.978L7.786 60.181L6.873 59.132Q6.737 58.985 6.588 58.953Q6.439 58.920 6.173 58.920L6.173 58.640L7.673 58.640L7.673 58.920Q7.482 58.920 7.482 59.054Q7.482 59.084 7.513 59.132L8.107 59.816L8.548 59.320Q8.661 59.190 8.661 59.074Q8.661 59.012 8.623 58.966Q8.586 58.920 8.528 58.920L8.528 58.640L9.844 58.640L9.844 58.920Q9.283 58.920 8.904 59.320L8.282 60.021L9.276 61.169Q9.375 61.268 9.476 61.313Q9.577 61.357 9.688 61.367Q9.799 61.378 9.977 61.378L9.977 61.658L8.483 61.658L8.483 61.378Q8.548 61.378 8.608 61.344Q8.668 61.309 8.668 61.244Q8.668 61.197 8.637 61.169L7.960 60.383L7.427 60.978Q7.314 61.108 7.314 61.224Q7.314 61.289 7.355 61.333Q7.396 61.378 7.455 61.378\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M108.28-54.998h56.057V-72.07H108.28Z\"\u002F>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(128.95 -123.442)\">\u003Cpath d=\"M-15.138 60.404L-17.196 60.404L-17.196 59.901L-15.138 59.901L-15.138 60.404M-12.667 61.658L-14.270 61.658L-14.270 61.378Q-14.044 61.378-13.896 61.344Q-13.747 61.309-13.747 61.169L-13.747 57.550Q-13.747 57.280-13.855 57.218Q-13.962 57.157-14.270 57.157L-14.270 56.876L-13.193 56.801L-13.193 61.169Q-13.193 61.306-13.043 61.342Q-12.893 61.378-12.667 61.378L-12.667 61.658M-12.113 60.175Q-12.113 59.833-11.978 59.534Q-11.843 59.235-11.604 59.011Q-11.365 58.787-11.047 58.662Q-10.729 58.537-10.398 58.537Q-9.953 58.537-9.553 58.753Q-9.153 58.968-8.919 59.346Q-8.685 59.723-8.685 60.175Q-8.685 60.516-8.827 60.800Q-8.969 61.084-9.213 61.291Q-9.458 61.497-9.767 61.612Q-10.076 61.726-10.398 61.726Q-10.828 61.726-11.230 61.525Q-11.631 61.323-11.872 60.971Q-12.113 60.619-12.113 60.175M-10.398 61.477Q-9.796 61.477-9.572 61.099Q-9.348 60.721-9.348 60.089Q-9.348 59.477-9.582 59.118Q-9.816 58.760-10.398 58.760Q-11.450 58.760-11.450 60.089Q-11.450 60.721-11.225 61.099Q-10.999 61.477-10.398 61.477M-8.131 62.191Q-8.131 61.945-7.935 61.761Q-7.738 61.576-7.482 61.497Q-7.619 61.385-7.690 61.224Q-7.762 61.063-7.762 60.882Q-7.762 60.561-7.550 60.315Q-7.885 60.017-7.885 59.607Q-7.885 59.146-7.496 58.859Q-7.106 58.572-6.628 58.572Q-6.156 58.572-5.821 58.818Q-5.647 58.664-5.436 58.582Q-5.226 58.500-4.997 58.500Q-4.833 58.500-4.712 58.607Q-4.590 58.715-4.590 58.879Q-4.590 58.975-4.662 59.047Q-4.734 59.118-4.826 59.118Q-4.925 59.118-4.995 59.045Q-5.065 58.971-5.065 58.872Q-5.065 58.818-5.052 58.787L-5.045 58.773Q-5.038 58.753-5.030 58.742Q-5.021 58.732-5.018 58.725Q-5.373 58.725-5.660 58.948Q-5.373 59.241-5.373 59.607Q-5.373 59.922-5.558 60.154Q-5.742 60.387-6.031 60.515Q-6.320 60.643-6.628 60.643Q-6.829 60.643-7.021 60.593Q-7.212 60.544-7.390 60.434Q-7.482 60.561-7.482 60.704Q-7.482 60.886-7.354 61.021Q-7.226 61.156-7.041 61.156L-6.409 61.156Q-5.961 61.156-5.592 61.227Q-5.223 61.299-4.963 61.528Q-4.703 61.757-4.703 62.191Q-4.703 62.512-4.999 62.714Q-5.294 62.916-5.698 63.005Q-6.101 63.094-6.416 63.094Q-6.733 63.094-7.137 63.005Q-7.540 62.916-7.836 62.714Q-8.131 62.512-8.131 62.191M-7.677 62.191Q-7.677 62.420-7.458 62.569Q-7.239 62.718-6.947 62.786Q-6.655 62.854-6.416 62.854Q-6.252 62.854-6.043 62.818Q-5.835 62.783-5.628 62.702Q-5.421 62.622-5.289 62.494Q-5.158 62.366-5.158 62.191Q-5.158 61.839-5.539 61.745Q-5.920 61.651-6.422 61.651L-7.041 61.651Q-7.280 61.651-7.479 61.802Q-7.677 61.952-7.677 62.191M-6.628 60.404Q-5.961 60.404-5.961 59.607Q-5.961 58.807-6.628 58.807Q-7.297 58.807-7.297 59.607Q-7.297 60.404-6.628 60.404\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(128.95 -123.442)\">\u003Cpath d=\"M-1.062 62.793Q-0.932 62.861-0.795 62.861Q-0.624 62.861-0.474 62.772Q-0.323 62.683-0.212 62.538Q-0.101 62.393-0.023 62.225L0.241 61.658L-0.928 59.132Q-1.003 58.985-1.133 58.953Q-1.263 58.920-1.496 58.920L-1.496 58.640L0.025 58.640L0.025 58.920Q-0.323 58.920-0.323 59.067Q-0.320 59.088-0.318 59.105Q-0.316 59.122-0.316 59.132L0.541 60.991L1.314 59.320Q1.348 59.252 1.348 59.173Q1.348 59.060 1.264 58.990Q1.181 58.920 1.068 58.920L1.068 58.640L2.264 58.640L2.264 58.920Q2.045 58.920 1.873 59.024Q1.700 59.129 1.608 59.320L0.271 62.225Q0.101 62.595-0.169 62.841Q-0.440 63.087-0.795 63.087Q-1.065 63.087-1.284 62.921Q-1.503 62.755-1.503 62.492Q-1.503 62.355-1.410 62.266Q-1.318 62.178-1.178 62.178Q-1.041 62.178-0.952 62.266Q-0.863 62.355-0.863 62.492Q-0.863 62.595-0.916 62.673Q-0.969 62.752-1.062 62.793M4.934 63.408Q4.383 63.008 4.012 62.453Q3.642 61.897 3.460 61.251Q3.279 60.605 3.279 59.908Q3.279 59.395 3.380 58.900Q3.481 58.404 3.686 57.953Q3.891 57.502 4.204 57.110Q4.517 56.719 4.934 56.415Q4.944 56.411 4.951 56.410Q4.957 56.408 4.968 56.408L5.036 56.408Q5.070 56.408 5.092 56.432Q5.115 56.456 5.115 56.493Q5.115 56.538 5.087 56.555Q4.739 56.856 4.486 57.240Q4.233 57.625 4.081 58.066Q3.929 58.507 3.857 58.963Q3.785 59.419 3.785 59.908Q3.785 60.909 4.094 61.796Q4.404 62.683 5.087 63.268Q5.115 63.285 5.115 63.329Q5.115 63.367 5.092 63.391Q5.070 63.415 5.036 63.415L4.968 63.415Q4.961 63.411 4.952 63.410Q4.944 63.408 4.934 63.408M6.451 60.817L6.451 58.920L5.812 58.920L5.812 58.698Q6.130 58.698 6.347 58.488Q6.564 58.278 6.665 57.968Q6.766 57.659 6.766 57.351L7.032 57.351L7.032 58.640L8.109 58.640L8.109 58.920L7.032 58.920L7.032 60.804Q7.032 61.080 7.136 61.279Q7.241 61.477 7.500 61.477Q7.658 61.477 7.764 61.373Q7.870 61.268 7.919 61.115Q7.969 60.961 7.969 60.804L7.969 60.390L8.235 60.390L8.235 60.817Q8.235 61.043 8.136 61.253Q8.037 61.463 7.852 61.595Q7.668 61.726 7.439 61.726Q7.001 61.726 6.726 61.489Q6.451 61.251 6.451 60.817M10.727 61.658L9.093 61.658L9.093 61.378Q9.322 61.378 9.471 61.344Q9.620 61.309 9.620 61.169L9.620 57.550Q9.620 57.280 9.512 57.218Q9.404 57.157 9.093 57.157L9.093 56.876L10.173 56.801L10.173 59.187Q10.279 59.002 10.457 58.860Q10.635 58.719 10.843 58.645Q11.052 58.572 11.277 58.572Q11.783 58.572 12.067 58.795Q12.351 59.019 12.351 59.515L12.351 61.169Q12.351 61.306 12.499 61.342Q12.648 61.378 12.873 61.378L12.873 61.658L11.243 61.658L11.243 61.378Q11.472 61.378 11.621 61.344Q11.769 61.309 11.769 61.169L11.769 59.529Q11.769 59.194 11.650 58.994Q11.530 58.794 11.216 58.794Q10.946 58.794 10.712 58.930Q10.477 59.067 10.339 59.301Q10.201 59.535 10.201 59.809L10.201 61.169Q10.201 61.306 10.351 61.342Q10.501 61.378 10.727 61.378L10.727 61.658M13.519 60.930Q13.519 60.598 13.743 60.371Q13.967 60.144 14.311 60.016Q14.654 59.887 15.027 59.835Q15.399 59.782 15.704 59.782L15.704 59.529Q15.704 59.324 15.596 59.144Q15.488 58.965 15.307 58.862Q15.126 58.760 14.917 58.760Q14.511 58.760 14.275 58.852Q14.364 58.889 14.410 58.973Q14.456 59.057 14.456 59.159Q14.456 59.255 14.410 59.334Q14.364 59.412 14.283 59.457Q14.203 59.501 14.114 59.501Q13.964 59.501 13.863 59.404Q13.762 59.306 13.762 59.159Q13.762 58.537 14.917 58.537Q15.129 58.537 15.379 58.601Q15.628 58.664 15.830 58.783Q16.032 58.903 16.158 59.088Q16.285 59.272 16.285 59.515L16.285 61.091Q16.285 61.207 16.346 61.303Q16.408 61.398 16.520 61.398Q16.630 61.398 16.695 61.304Q16.760 61.210 16.760 61.091L16.760 60.643L17.026 60.643L17.026 61.091Q17.026 61.361 16.799 61.526Q16.572 61.692 16.291 61.692Q16.083 61.692 15.946 61.538Q15.810 61.385 15.786 61.169Q15.639 61.436 15.357 61.581Q15.075 61.726 14.750 61.726Q14.473 61.726 14.189 61.651Q13.906 61.576 13.713 61.397Q13.519 61.217 13.519 60.930M14.135 60.930Q14.135 61.104 14.236 61.234Q14.336 61.364 14.492 61.434Q14.647 61.504 14.811 61.504Q15.030 61.504 15.239 61.407Q15.447 61.309 15.575 61.128Q15.704 60.947 15.704 60.721L15.704 59.993Q15.379 59.993 15.013 60.084Q14.647 60.175 14.391 60.387Q14.135 60.598 14.135 60.930M19.125 61.658L17.491 61.658L17.491 61.378Q17.720 61.378 17.869 61.344Q18.018 61.309 18.018 61.169L18.018 59.320Q18.018 59.050 17.910 58.989Q17.802 58.927 17.491 58.927L17.491 58.647L18.551 58.572L18.551 59.221Q18.722 58.913 19.026 58.742Q19.330 58.572 19.675 58.572Q20.181 58.572 20.465 58.795Q20.748 59.019 20.748 59.515L20.748 61.169Q20.748 61.306 20.897 61.342Q21.046 61.378 21.271 61.378L21.271 61.658L19.641 61.658L19.641 61.378Q19.870 61.378 20.019 61.344Q20.167 61.309 20.167 61.169L20.167 59.529Q20.167 59.194 20.048 58.994Q19.928 58.794 19.614 58.794Q19.344 58.794 19.110 58.930Q18.875 59.067 18.737 59.301Q18.599 59.535 18.599 59.809L18.599 61.169Q18.599 61.306 18.749 61.342Q18.899 61.378 19.125 61.378L19.125 61.658M23.456 61.658L21.873 61.658L21.873 61.378Q22.102 61.378 22.251 61.344Q22.399 61.309 22.399 61.169L22.399 57.550Q22.399 57.280 22.292 57.218Q22.184 57.157 21.873 57.157L21.873 56.876L22.953 56.801L22.953 60.089L23.937 59.320Q24.143 59.183 24.143 59.033Q24.143 58.989 24.102 58.954Q24.060 58.920 24.016 58.920L24.016 58.640L25.380 58.640L25.380 58.920Q24.891 58.920 24.372 59.320L23.814 59.754L24.792 60.978Q24.994 61.224 25.127 61.301Q25.260 61.378 25.547 61.378L25.547 61.658L24.115 61.658L24.115 61.378Q24.303 61.378 24.303 61.265Q24.303 61.169 24.149 60.978L23.414 60.069L22.933 60.448L22.933 61.169Q22.933 61.306 23.081 61.342Q23.230 61.378 23.456 61.378L23.456 61.658M26.060 61.651L26.060 60.588Q26.060 60.564 26.087 60.537Q26.115 60.510 26.139 60.510L26.248 60.510Q26.313 60.510 26.327 60.568Q26.422 61.002 26.668 61.253Q26.914 61.504 27.328 61.504Q27.670 61.504 27.923 61.371Q28.176 61.238 28.176 60.930Q28.176 60.773 28.082 60.658Q27.988 60.544 27.849 60.475Q27.711 60.407 27.543 60.369L26.962 60.270Q26.607 60.202 26.333 59.981Q26.060 59.761 26.060 59.419Q26.060 59.170 26.171 58.995Q26.282 58.821 26.468 58.722Q26.655 58.623 26.870 58.580Q27.085 58.537 27.328 58.537Q27.742 58.537 28.022 58.719L28.237 58.544Q28.247 58.541 28.254 58.539Q28.261 58.537 28.271 58.537L28.323 58.537Q28.350 58.537 28.374 58.561Q28.398 58.585 28.398 58.613L28.398 59.460Q28.398 59.481 28.374 59.508Q28.350 59.535 28.323 59.535L28.210 59.535Q28.183 59.535 28.157 59.510Q28.131 59.484 28.131 59.460Q28.131 59.224 28.025 59.060Q27.919 58.896 27.737 58.814Q27.554 58.732 27.321 58.732Q26.993 58.732 26.737 58.835Q26.480 58.937 26.480 59.214Q26.480 59.409 26.663 59.518Q26.846 59.628 27.075 59.669L27.649 59.775Q27.895 59.823 28.109 59.951Q28.323 60.079 28.459 60.282Q28.596 60.486 28.596 60.735Q28.596 61.248 28.230 61.487Q27.865 61.726 27.328 61.726Q26.832 61.726 26.501 61.432L26.234 61.706Q26.214 61.726 26.186 61.726L26.139 61.726Q26.115 61.726 26.087 61.699Q26.060 61.672 26.060 61.651M29.546 63.415L29.478 63.415Q29.444 63.415 29.422 63.389Q29.399 63.364 29.399 63.329Q29.399 63.285 29.430 63.268Q29.786 62.964 30.035 62.574Q30.285 62.184 30.437 61.752Q30.589 61.320 30.659 60.851Q30.729 60.383 30.729 59.908Q30.729 59.429 30.659 58.963Q30.589 58.496 30.435 58.061Q30.281 57.625 30.030 57.237Q29.779 56.849 29.430 56.555Q29.399 56.538 29.399 56.493Q29.399 56.459 29.422 56.434Q29.444 56.408 29.478 56.408L29.546 56.408Q29.557 56.408 29.565 56.410Q29.574 56.411 29.584 56.415Q30.127 56.815 30.500 57.368Q30.872 57.922 31.054 58.568Q31.235 59.214 31.235 59.908Q31.235 60.609 31.054 61.256Q30.872 61.904 30.498 62.458Q30.124 63.012 29.584 63.408Q29.574 63.408 29.565 63.410Q29.557 63.411 29.546 63.415\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M136.308 52.922V33.36\"\u002F>\u003Cpath stroke=\"none\" d=\"m136.308 30.76-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M136.308 7.198v-19.563\"\u002F>\u003Cpath stroke=\"none\" d=\"m136.308-14.965-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M136.308-32.436v-19.762\"\u002F>\u003Cpath stroke=\"none\" d=\"m136.308-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M198.14 70.194h29.98V53.122h-29.98Z\"\u002F>\u003Cg transform=\"translate(218.81 2.43)\">\u003Cpath d=\"M-16.495 60.817L-16.495 58.920L-17.134 58.920L-17.134 58.698Q-16.816 58.698-16.599 58.488Q-16.382 58.278-16.282 57.968Q-16.181 57.659-16.181 57.351L-15.914 57.351L-15.914 58.640L-14.837 58.640L-14.837 58.920L-15.914 58.920L-15.914 60.804Q-15.914 61.080-15.810 61.279Q-15.706 61.477-15.446 61.477Q-15.289 61.477-15.183 61.373Q-15.077 61.268-15.027 61.115Q-14.978 60.961-14.978 60.804L-14.978 60.390L-14.711 60.390L-14.711 60.817Q-14.711 61.043-14.810 61.253Q-14.909 61.463-15.094 61.595Q-15.278 61.726-15.507 61.726Q-15.945 61.726-16.220 61.489Q-16.495 61.251-16.495 60.817M-12.219 61.658L-13.853 61.658L-13.853 61.378Q-13.624 61.378-13.475 61.344Q-13.327 61.309-13.327 61.169L-13.327 57.550Q-13.327 57.280-13.434 57.218Q-13.542 57.157-13.853 57.157L-13.853 56.876L-12.773 56.801L-12.773 59.187Q-12.667 59.002-12.489 58.860Q-12.312 58.719-12.103 58.645Q-11.895 58.572-11.669 58.572Q-11.163 58.572-10.879 58.795Q-10.596 59.019-10.596 59.515L-10.596 61.169Q-10.596 61.306-10.447 61.342Q-10.298 61.378-10.073 61.378L-10.073 61.658L-11.703 61.658L-11.703 61.378Q-11.474 61.378-11.326 61.344Q-11.177 61.309-11.177 61.169L-11.177 59.529Q-11.177 59.194-11.296 58.994Q-11.416 58.794-11.731 58.794Q-12.001 58.794-12.235 58.930Q-12.469 59.067-12.607 59.301Q-12.746 59.535-12.746 59.809L-12.746 61.169Q-12.746 61.306-12.595 61.342Q-12.445 61.378-12.219 61.378L-12.219 61.658M-9.427 60.930Q-9.427 60.598-9.203 60.371Q-8.979 60.144-8.636 60.016Q-8.292 59.887-7.919 59.835Q-7.547 59.782-7.243 59.782L-7.243 59.529Q-7.243 59.324-7.350 59.144Q-7.458 58.965-7.639 58.862Q-7.820 58.760-8.029 58.760Q-8.436 58.760-8.671 58.852Q-8.583 58.889-8.536 58.973Q-8.490 59.057-8.490 59.159Q-8.490 59.255-8.536 59.334Q-8.583 59.412-8.663 59.457Q-8.743 59.501-8.832 59.501Q-8.982 59.501-9.083 59.404Q-9.184 59.306-9.184 59.159Q-9.184 58.537-8.029 58.537Q-7.817 58.537-7.567 58.601Q-7.318 58.664-7.116 58.783Q-6.915 58.903-6.788 59.088Q-6.662 59.272-6.662 59.515L-6.662 61.091Q-6.662 61.207-6.600 61.303Q-6.539 61.398-6.426 61.398Q-6.316 61.398-6.252 61.304Q-6.187 61.210-6.187 61.091L-6.187 60.643L-5.920 60.643L-5.920 61.091Q-5.920 61.361-6.147 61.526Q-6.375 61.692-6.655 61.692Q-6.863 61.692-7 61.538Q-7.137 61.385-7.161 61.169Q-7.308 61.436-7.590 61.581Q-7.872 61.726-8.196 61.726Q-8.473 61.726-8.757 61.651Q-9.041 61.576-9.234 61.397Q-9.427 61.217-9.427 60.930M-8.812 60.930Q-8.812 61.104-8.711 61.234Q-8.610 61.364-8.454 61.434Q-8.299 61.504-8.135 61.504Q-7.916 61.504-7.708 61.407Q-7.499 61.309-7.371 61.128Q-7.243 60.947-7.243 60.721L-7.243 59.993Q-7.567 59.993-7.933 60.084Q-8.299 60.175-8.555 60.387Q-8.812 60.598-8.812 60.930M-3.821 61.658L-5.455 61.658L-5.455 61.378Q-5.226 61.378-5.077 61.344Q-4.929 61.309-4.929 61.169L-4.929 59.320Q-4.929 59.050-5.036 58.989Q-5.144 58.927-5.455 58.927L-5.455 58.647L-4.396 58.572L-4.396 59.221Q-4.225 58.913-3.920 58.742Q-3.616 58.572-3.271 58.572Q-2.765 58.572-2.482 58.795Q-2.198 59.019-2.198 59.515L-2.198 61.169Q-2.198 61.306-2.049 61.342Q-1.900 61.378-1.675 61.378L-1.675 61.658L-3.305 61.658L-3.305 61.378Q-3.076 61.378-2.928 61.344Q-2.779 61.309-2.779 61.169L-2.779 59.529Q-2.779 59.194-2.898 58.994Q-3.018 58.794-3.333 58.794Q-3.603 58.794-3.837 58.930Q-4.071 59.067-4.209 59.301Q-4.348 59.535-4.348 59.809L-4.348 61.169Q-4.348 61.306-4.197 61.342Q-4.047 61.378-3.821 61.378L-3.821 61.658M0.509 61.658L-1.073 61.658L-1.073 61.378Q-0.844 61.378-0.696 61.344Q-0.547 61.309-0.547 61.169L-0.547 57.550Q-0.547 57.280-0.655 57.218Q-0.762 57.157-1.073 57.157L-1.073 56.876L0.007 56.801L0.007 60.089L0.991 59.320Q1.196 59.183 1.196 59.033Q1.196 58.989 1.155 58.954Q1.114 58.920 1.070 58.920L1.070 58.640L2.434 58.640L2.434 58.920Q1.945 58.920 1.425 59.320L0.868 59.754L1.846 60.978Q2.047 61.224 2.181 61.301Q2.314 61.378 2.601 61.378L2.601 61.658L1.169 61.658L1.169 61.378Q1.357 61.378 1.357 61.265Q1.357 61.169 1.203 60.978L0.468 60.069L-0.014 60.448L-0.014 61.169Q-0.014 61.306 0.135 61.342Q0.284 61.378 0.509 61.378L0.509 61.658M3.114 61.651L3.114 60.588Q3.114 60.564 3.141 60.537Q3.168 60.510 3.192 60.510L3.302 60.510Q3.367 60.510 3.380 60.568Q3.476 61.002 3.722 61.253Q3.968 61.504 4.382 61.504Q4.724 61.504 4.977 61.371Q5.229 61.238 5.229 60.930Q5.229 60.773 5.135 60.658Q5.041 60.544 4.903 60.475Q4.765 60.407 4.597 60.369L4.016 60.270Q3.661 60.202 3.387 59.981Q3.114 59.761 3.114 59.419Q3.114 59.170 3.225 58.995Q3.336 58.821 3.522 58.722Q3.708 58.623 3.924 58.580Q4.139 58.537 4.382 58.537Q4.795 58.537 5.076 58.719L5.291 58.544Q5.301 58.541 5.308 58.539Q5.315 58.537 5.325 58.537L5.376 58.537Q5.404 58.537 5.428 58.561Q5.452 58.585 5.452 58.613L5.452 59.460Q5.452 59.481 5.428 59.508Q5.404 59.535 5.376 59.535L5.264 59.535Q5.236 59.535 5.211 59.510Q5.185 59.484 5.185 59.460Q5.185 59.224 5.079 59.060Q4.973 58.896 4.790 58.814Q4.607 58.732 4.375 58.732Q4.047 58.732 3.790 58.835Q3.534 58.937 3.534 59.214Q3.534 59.409 3.717 59.518Q3.900 59.628 4.129 59.669L4.703 59.775Q4.949 59.823 5.163 59.951Q5.376 60.079 5.513 60.282Q5.650 60.486 5.650 60.735Q5.650 61.248 5.284 61.487Q4.918 61.726 4.382 61.726Q3.886 61.726 3.555 61.432L3.288 61.706Q3.268 61.726 3.240 61.726L3.192 61.726Q3.168 61.726 3.141 61.699Q3.114 61.672 3.114 61.651\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M200.327 30.36h25.608V7.597h-25.608Z\"\u002F>\u003Cg transform=\"translate(227.898 -39.554)\">\u003Cpath d=\"M-14.950 61.658L-17.037 61.658L-17.037 61.342Q-16.730 61.342-16.538 61.289Q-16.347 61.236-16.347 61.047L-16.347 56.332Q-16.347 56.090-16.418 55.982Q-16.488 55.875-16.622 55.851Q-16.756 55.826-17.037 55.826L-17.037 55.510L-15.670 55.413L-15.670 58.463Q-15.473 58.107-15.119 57.894Q-14.765 57.681-14.365 57.681Q-13.086 57.681-13.086 58.894L-13.086 61.047Q-13.086 61.236-12.895 61.289Q-12.704 61.342-12.397 61.342L-12.397 61.658L-14.484 61.658L-14.484 61.342Q-14.172 61.342-13.981 61.289Q-13.790 61.236-13.790 61.047L-13.790 58.929Q-13.790 58.670-13.834 58.448Q-13.878 58.226-14.023 58.083Q-14.168 57.940-14.427 57.940Q-14.770 57.940-15.051 58.129Q-15.332 58.318-15.488 58.630Q-15.644 58.942-15.644 59.289L-15.644 61.047Q-15.644 61.236-15.451 61.289Q-15.257 61.342-14.950 61.342\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.270\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M196.04-15.165h34.181v-17.071h-34.18Z\"\u002F>\u003Cg transform=\"translate(216.71 -82.928)\">\u003Cpath d=\"M-17.022 61.651L-17.022 60.588Q-17.022 60.564-16.994 60.537Q-16.967 60.510-16.943 60.510L-16.834 60.510Q-16.769 60.510-16.755 60.568Q-16.659 61.002-16.413 61.253Q-16.167 61.504-15.753 61.504Q-15.412 61.504-15.159 61.371Q-14.906 61.238-14.906 60.930Q-14.906 60.773-15 60.658Q-15.094 60.544-15.232 60.475Q-15.371 60.407-15.538 60.369L-16.119 60.270Q-16.475 60.202-16.748 59.981Q-17.022 59.761-17.022 59.419Q-17.022 59.170-16.910 58.995Q-16.799 58.821-16.613 58.722Q-16.427 58.623-16.211 58.580Q-15.996 58.537-15.753 58.537Q-15.340 58.537-15.060 58.719L-14.844 58.544Q-14.834 58.541-14.827 58.539Q-14.820 58.537-14.810 58.537L-14.759 58.537Q-14.732 58.537-14.708 58.561Q-14.684 58.585-14.684 58.613L-14.684 59.460Q-14.684 59.481-14.708 59.508Q-14.732 59.535-14.759 59.535L-14.872 59.535Q-14.899 59.535-14.925 59.510Q-14.950 59.484-14.950 59.460Q-14.950 59.224-15.056 59.060Q-15.162 58.896-15.345 58.814Q-15.528 58.732-15.760 58.732Q-16.088 58.732-16.345 58.835Q-16.601 58.937-16.601 59.214Q-16.601 59.409-16.418 59.518Q-16.235 59.628-16.006 59.669L-15.432 59.775Q-15.186 59.823-14.972 59.951Q-14.759 60.079-14.622 60.282Q-14.485 60.486-14.485 60.735Q-14.485 61.248-14.851 61.487Q-15.217 61.726-15.753 61.726Q-16.249 61.726-16.581 61.432L-16.847 61.706Q-16.868 61.726-16.895 61.726L-16.943 61.726Q-16.967 61.726-16.994 61.699Q-17.022 61.672-17.022 61.651M-13.898 60.175Q-13.898 59.833-13.763 59.534Q-13.628 59.235-13.388 59.011Q-13.149 58.787-12.831 58.662Q-12.513 58.537-12.182 58.537Q-11.737 58.537-11.337 58.753Q-10.938 58.968-10.703 59.346Q-10.469 59.723-10.469 60.175Q-10.469 60.516-10.611 60.800Q-10.753 61.084-10.997 61.291Q-11.242 61.497-11.551 61.612Q-11.860 61.726-12.182 61.726Q-12.612 61.726-13.014 61.525Q-13.416 61.323-13.657 60.971Q-13.898 60.619-13.898 60.175M-12.182 61.477Q-11.580 61.477-11.356 61.099Q-11.132 60.721-11.132 60.089Q-11.132 59.477-11.367 59.118Q-11.601 58.760-12.182 58.760Q-13.234 58.760-13.234 60.089Q-13.234 60.721-13.009 61.099Q-12.783 61.477-12.182 61.477M-8.077 61.658L-9.810 61.658L-9.810 61.378Q-9.584 61.378-9.435 61.344Q-9.287 61.309-9.287 61.169L-9.287 58.920L-9.875 58.920L-9.875 58.640L-9.287 58.640L-9.287 57.823Q-9.287 57.505-9.109 57.257Q-8.931 57.010-8.641 56.869Q-8.350 56.729-8.039 56.729Q-7.783 56.729-7.579 56.871Q-7.376 57.013-7.376 57.256Q-7.376 57.392-7.475 57.491Q-7.574 57.591-7.711 57.591Q-7.848 57.591-7.947 57.491Q-8.046 57.392-8.046 57.256Q-8.046 57.075-7.906 56.982Q-7.984 56.955-8.084 56.955Q-8.292 56.955-8.446 57.088Q-8.600 57.221-8.680 57.425Q-8.760 57.628-8.760 57.837L-8.760 58.640L-7.872 58.640L-7.872 58.920L-8.733 58.920L-8.733 61.169Q-8.733 61.378-8.077 61.378L-8.077 61.658M-6.870 60.817L-6.870 58.920L-7.509 58.920L-7.509 58.698Q-7.191 58.698-6.974 58.488Q-6.757 58.278-6.657 57.968Q-6.556 57.659-6.556 57.351L-6.289 57.351L-6.289 58.640L-5.212 58.640L-5.212 58.920L-6.289 58.920L-6.289 60.804Q-6.289 61.080-6.185 61.279Q-6.081 61.477-5.821 61.477Q-5.664 61.477-5.558 61.373Q-5.452 61.268-5.402 61.115Q-5.353 60.961-5.353 60.804L-5.353 60.390L-5.086 60.390L-5.086 60.817Q-5.086 61.043-5.185 61.253Q-5.284 61.463-5.469 61.595Q-5.653 61.726-5.882 61.726Q-6.320 61.726-6.595 61.489Q-6.870 61.251-6.870 60.817M-2.594 61.658L-4.228 61.658L-4.228 61.378Q-3.999 61.378-3.850 61.344Q-3.702 61.309-3.702 61.169L-3.702 59.320Q-3.702 59.050-3.809 58.989Q-3.917 58.927-4.228 58.927L-4.228 58.647L-3.169 58.572L-3.169 59.221Q-2.998 58.913-2.693 58.742Q-2.389 58.572-2.044 58.572Q-1.644 58.572-1.367 58.712Q-1.090 58.852-1.005 59.200Q-0.837 58.907-0.538 58.739Q-0.239 58.572 0.106 58.572Q0.612 58.572 0.895 58.795Q1.179 59.019 1.179 59.515L1.179 61.169Q1.179 61.306 1.328 61.342Q1.477 61.378 1.702 61.378L1.702 61.658L0.072 61.658L0.072 61.378Q0.297 61.378 0.448 61.342Q0.598 61.306 0.598 61.169L0.598 59.529Q0.598 59.194 0.478 58.994Q0.359 58.794 0.044 58.794Q-0.226 58.794-0.460 58.930Q-0.694 59.067-0.832 59.301Q-0.971 59.535-0.971 59.809L-0.971 61.169Q-0.971 61.306-0.822 61.342Q-0.673 61.378-0.448 61.378L-0.448 61.658L-2.078 61.658L-2.078 61.378Q-1.849 61.378-1.701 61.344Q-1.552 61.309-1.552 61.169L-1.552 59.529Q-1.552 59.194-1.671 58.994Q-1.791 58.794-2.106 58.794Q-2.376 58.794-2.610 58.930Q-2.844 59.067-2.982 59.301Q-3.121 59.535-3.121 59.809L-3.121 61.169Q-3.121 61.306-2.970 61.342Q-2.820 61.378-2.594 61.378L-2.594 61.658M2.348 60.930Q2.348 60.598 2.572 60.371Q2.796 60.144 3.139 60.016Q3.483 59.887 3.855 59.835Q4.228 59.782 4.532 59.782L4.532 59.529Q4.532 59.324 4.424 59.144Q4.317 58.965 4.136 58.862Q3.955 58.760 3.746 58.760Q3.339 58.760 3.103 58.852Q3.192 58.889 3.238 58.973Q3.285 59.057 3.285 59.159Q3.285 59.255 3.238 59.334Q3.192 59.412 3.112 59.457Q3.032 59.501 2.943 59.501Q2.792 59.501 2.692 59.404Q2.591 59.306 2.591 59.159Q2.591 58.537 3.746 58.537Q3.958 58.537 4.207 58.601Q4.457 58.664 4.659 58.783Q4.860 58.903 4.987 59.088Q5.113 59.272 5.113 59.515L5.113 61.091Q5.113 61.207 5.175 61.303Q5.236 61.398 5.349 61.398Q5.458 61.398 5.523 61.304Q5.588 61.210 5.588 61.091L5.588 60.643L5.855 60.643L5.855 61.091Q5.855 61.361 5.628 61.526Q5.400 61.692 5.120 61.692Q4.912 61.692 4.775 61.538Q4.638 61.385 4.614 61.169Q4.467 61.436 4.185 61.581Q3.903 61.726 3.579 61.726Q3.302 61.726 3.018 61.651Q2.734 61.576 2.541 61.397Q2.348 61.217 2.348 60.930M2.963 60.930Q2.963 61.104 3.064 61.234Q3.165 61.364 3.320 61.434Q3.476 61.504 3.640 61.504Q3.859 61.504 4.067 61.407Q4.276 61.309 4.404 61.128Q4.532 60.947 4.532 60.721L4.532 59.993Q4.207 59.993 3.842 60.084Q3.476 60.175 3.220 60.387Q2.963 60.598 2.963 60.930M7.455 61.658L6.132 61.658L6.132 61.378Q6.692 61.378 7.072 60.978L7.786 60.181L6.873 59.132Q6.737 58.985 6.588 58.953Q6.439 58.920 6.173 58.920L6.173 58.640L7.673 58.640L7.673 58.920Q7.482 58.920 7.482 59.054Q7.482 59.084 7.513 59.132L8.107 59.816L8.548 59.320Q8.661 59.190 8.661 59.074Q8.661 59.012 8.623 58.966Q8.586 58.920 8.528 58.920L8.528 58.640L9.844 58.640L9.844 58.920Q9.283 58.920 8.904 59.320L8.282 60.021L9.276 61.169Q9.375 61.268 9.476 61.313Q9.577 61.357 9.688 61.367Q9.799 61.378 9.977 61.378L9.977 61.658L8.483 61.658L8.483 61.378Q8.548 61.378 8.608 61.344Q8.668 61.309 8.668 61.244Q8.668 61.197 8.637 61.169L7.960 60.383L7.427 60.978Q7.314 61.108 7.314 61.224Q7.314 61.289 7.355 61.333Q7.396 61.378 7.455 61.378\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M191.964-54.998h42.333V-72.07h-42.333Z\"\u002F>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(212.634 -123.442)\">\u003Cpath d=\"M-15.138 60.404L-17.196 60.404L-17.196 59.901L-15.138 59.901L-15.138 60.404M-12.667 61.658L-14.270 61.658L-14.270 61.378Q-14.044 61.378-13.896 61.344Q-13.747 61.309-13.747 61.169L-13.747 57.550Q-13.747 57.280-13.855 57.218Q-13.962 57.157-14.270 57.157L-14.270 56.876L-13.193 56.801L-13.193 61.169Q-13.193 61.306-13.043 61.342Q-12.893 61.378-12.667 61.378L-12.667 61.658M-12.113 60.175Q-12.113 59.833-11.978 59.534Q-11.843 59.235-11.604 59.011Q-11.365 58.787-11.047 58.662Q-10.729 58.537-10.398 58.537Q-9.953 58.537-9.553 58.753Q-9.153 58.968-8.919 59.346Q-8.685 59.723-8.685 60.175Q-8.685 60.516-8.827 60.800Q-8.969 61.084-9.213 61.291Q-9.458 61.497-9.767 61.612Q-10.076 61.726-10.398 61.726Q-10.828 61.726-11.230 61.525Q-11.631 61.323-11.872 60.971Q-12.113 60.619-12.113 60.175M-10.398 61.477Q-9.796 61.477-9.572 61.099Q-9.348 60.721-9.348 60.089Q-9.348 59.477-9.582 59.118Q-9.816 58.760-10.398 58.760Q-11.450 58.760-11.450 60.089Q-11.450 60.721-11.225 61.099Q-10.999 61.477-10.398 61.477M-8.131 62.191Q-8.131 61.945-7.935 61.761Q-7.738 61.576-7.482 61.497Q-7.619 61.385-7.690 61.224Q-7.762 61.063-7.762 60.882Q-7.762 60.561-7.550 60.315Q-7.885 60.017-7.885 59.607Q-7.885 59.146-7.496 58.859Q-7.106 58.572-6.628 58.572Q-6.156 58.572-5.821 58.818Q-5.647 58.664-5.436 58.582Q-5.226 58.500-4.997 58.500Q-4.833 58.500-4.712 58.607Q-4.590 58.715-4.590 58.879Q-4.590 58.975-4.662 59.047Q-4.734 59.118-4.826 59.118Q-4.925 59.118-4.995 59.045Q-5.065 58.971-5.065 58.872Q-5.065 58.818-5.052 58.787L-5.045 58.773Q-5.038 58.753-5.030 58.742Q-5.021 58.732-5.018 58.725Q-5.373 58.725-5.660 58.948Q-5.373 59.241-5.373 59.607Q-5.373 59.922-5.558 60.154Q-5.742 60.387-6.031 60.515Q-6.320 60.643-6.628 60.643Q-6.829 60.643-7.021 60.593Q-7.212 60.544-7.390 60.434Q-7.482 60.561-7.482 60.704Q-7.482 60.886-7.354 61.021Q-7.226 61.156-7.041 61.156L-6.409 61.156Q-5.961 61.156-5.592 61.227Q-5.223 61.299-4.963 61.528Q-4.703 61.757-4.703 62.191Q-4.703 62.512-4.999 62.714Q-5.294 62.916-5.698 63.005Q-6.101 63.094-6.416 63.094Q-6.733 63.094-7.137 63.005Q-7.540 62.916-7.836 62.714Q-8.131 62.512-8.131 62.191M-7.677 62.191Q-7.677 62.420-7.458 62.569Q-7.239 62.718-6.947 62.786Q-6.655 62.854-6.416 62.854Q-6.252 62.854-6.043 62.818Q-5.835 62.783-5.628 62.702Q-5.421 62.622-5.289 62.494Q-5.158 62.366-5.158 62.191Q-5.158 61.839-5.539 61.745Q-5.920 61.651-6.422 61.651L-7.041 61.651Q-7.280 61.651-7.479 61.802Q-7.677 61.952-7.677 62.191M-6.628 60.404Q-5.961 60.404-5.961 59.607Q-5.961 58.807-6.628 58.807Q-7.297 58.807-7.297 59.607Q-7.297 60.404-6.628 60.404\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(212.634 -123.442)\">\u003Cpath d=\"M-1.062 62.793Q-0.932 62.861-0.795 62.861Q-0.624 62.861-0.474 62.772Q-0.323 62.683-0.212 62.538Q-0.101 62.393-0.023 62.225L0.241 61.658L-0.928 59.132Q-1.003 58.985-1.133 58.953Q-1.263 58.920-1.496 58.920L-1.496 58.640L0.025 58.640L0.025 58.920Q-0.323 58.920-0.323 59.067Q-0.320 59.088-0.318 59.105Q-0.316 59.122-0.316 59.132L0.541 60.991L1.314 59.320Q1.348 59.252 1.348 59.173Q1.348 59.060 1.264 58.990Q1.181 58.920 1.068 58.920L1.068 58.640L2.264 58.640L2.264 58.920Q2.045 58.920 1.873 59.024Q1.700 59.129 1.608 59.320L0.271 62.225Q0.101 62.595-0.169 62.841Q-0.440 63.087-0.795 63.087Q-1.065 63.087-1.284 62.921Q-1.503 62.755-1.503 62.492Q-1.503 62.355-1.410 62.266Q-1.318 62.178-1.178 62.178Q-1.041 62.178-0.952 62.266Q-0.863 62.355-0.863 62.492Q-0.863 62.595-0.916 62.673Q-0.969 62.752-1.062 62.793M4.934 63.408Q4.383 63.008 4.012 62.453Q3.642 61.897 3.460 61.251Q3.279 60.605 3.279 59.908Q3.279 59.395 3.380 58.900Q3.481 58.404 3.686 57.953Q3.891 57.502 4.204 57.110Q4.517 56.719 4.934 56.415Q4.944 56.411 4.951 56.410Q4.957 56.408 4.968 56.408L5.036 56.408Q5.070 56.408 5.092 56.432Q5.115 56.456 5.115 56.493Q5.115 56.538 5.087 56.555Q4.739 56.856 4.486 57.240Q4.233 57.625 4.081 58.066Q3.929 58.507 3.857 58.963Q3.785 59.419 3.785 59.908Q3.785 60.909 4.094 61.796Q4.404 62.683 5.087 63.268Q5.115 63.285 5.115 63.329Q5.115 63.367 5.092 63.391Q5.070 63.415 5.036 63.415L4.968 63.415Q4.961 63.411 4.952 63.410Q4.944 63.408 4.934 63.408M7.723 61.658L5.990 61.658L5.990 61.378Q6.215 61.378 6.364 61.344Q6.513 61.309 6.513 61.169L6.513 58.920L5.925 58.920L5.925 58.640L6.513 58.640L6.513 57.823Q6.513 57.505 6.690 57.257Q6.868 57.010 7.159 56.869Q7.449 56.729 7.760 56.729Q8.017 56.729 8.220 56.871Q8.423 57.013 8.423 57.256Q8.423 57.392 8.324 57.491Q8.225 57.591 8.088 57.591Q7.952 57.591 7.852 57.491Q7.753 57.392 7.753 57.256Q7.753 57.075 7.893 56.982Q7.815 56.955 7.716 56.955Q7.507 56.955 7.353 57.088Q7.200 57.221 7.119 57.425Q7.039 57.628 7.039 57.837L7.039 58.640L7.928 58.640L7.928 58.920L7.066 58.920L7.066 61.169Q7.066 61.378 7.723 61.378L7.723 61.658M8.362 60.175Q8.362 59.833 8.497 59.534Q8.632 59.235 8.871 59.011Q9.110 58.787 9.428 58.662Q9.746 58.537 10.078 58.537Q10.522 58.537 10.922 58.753Q11.322 58.968 11.556 59.346Q11.790 59.723 11.790 60.175Q11.790 60.516 11.648 60.800Q11.506 61.084 11.262 61.291Q11.018 61.497 10.708 61.612Q10.399 61.726 10.078 61.726Q9.647 61.726 9.245 61.525Q8.844 61.323 8.603 60.971Q8.362 60.619 8.362 60.175M10.078 61.477Q10.679 61.477 10.903 61.099Q11.127 60.721 11.127 60.089Q11.127 59.477 10.893 59.118Q10.659 58.760 10.078 58.760Q9.025 58.760 9.025 60.089Q9.025 60.721 9.250 61.099Q9.476 61.477 10.078 61.477M14.135 61.658L12.398 61.658L12.398 61.378Q12.627 61.378 12.776 61.344Q12.925 61.309 12.925 61.169L12.925 59.320Q12.925 59.050 12.817 58.989Q12.709 58.927 12.398 58.927L12.398 58.647L13.427 58.572L13.427 59.279Q13.557 58.971 13.800 58.772Q14.042 58.572 14.360 58.572Q14.579 58.572 14.750 58.696Q14.921 58.821 14.921 59.033Q14.921 59.170 14.822 59.269Q14.723 59.368 14.589 59.368Q14.453 59.368 14.353 59.269Q14.254 59.170 14.254 59.033Q14.254 58.893 14.353 58.794Q14.063 58.794 13.863 58.990Q13.663 59.187 13.571 59.481Q13.478 59.775 13.478 60.055L13.478 61.169Q13.478 61.378 14.135 61.378L14.135 61.658M15.827 63.415L15.758 63.415Q15.724 63.415 15.702 63.389Q15.680 63.364 15.680 63.329Q15.680 63.285 15.710 63.268Q16.066 62.964 16.315 62.574Q16.565 62.184 16.717 61.752Q16.869 61.320 16.939 60.851Q17.009 60.383 17.009 59.908Q17.009 59.429 16.939 58.963Q16.869 58.496 16.715 58.061Q16.561 57.625 16.310 57.237Q16.059 56.849 15.710 56.555Q15.680 56.538 15.680 56.493Q15.680 56.459 15.702 56.434Q15.724 56.408 15.758 56.408L15.827 56.408Q15.837 56.408 15.845 56.410Q15.854 56.411 15.864 56.415Q16.408 56.815 16.780 57.368Q17.153 57.922 17.334 58.568Q17.515 59.214 17.515 59.908Q17.515 60.609 17.334 61.256Q17.153 61.904 16.779 62.458Q16.404 63.012 15.864 63.408Q15.854 63.408 15.845 63.410Q15.837 63.411 15.827 63.415\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M213.13 52.922V33.36\"\u002F>\u003Cpath stroke=\"none\" d=\"m213.13 30.76-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M213.13 7.198v-19.563\"\u002F>\u003Cpath stroke=\"none\" d=\"m213.13-14.965-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M213.13-32.436v-19.762\"\u002F>\u003Cpath stroke=\"none\" d=\"m213.13-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-4.132 18.979h47.814\"\u002F>\u003Cpath stroke=\"none\" d=\"m46.282 18.979-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003Cg transform=\"translate(36.203 -46.412)\">\u003Cpath d=\"M-15.340 61.658L-16.974 61.658L-16.974 61.378Q-16.745 61.378-16.596 61.344Q-16.447 61.309-16.447 61.169L-16.447 57.550Q-16.447 57.280-16.555 57.218Q-16.663 57.157-16.974 57.157L-16.974 56.876L-15.894 56.801L-15.894 59.187Q-15.788 59.002-15.610 58.860Q-15.432 58.719-15.224 58.645Q-15.015 58.572-14.790 58.572Q-14.284 58.572-14 58.795Q-13.716 59.019-13.716 59.515L-13.716 61.169Q-13.716 61.306-13.568 61.342Q-13.419 61.378-13.193 61.378L-13.193 61.658L-14.824 61.658L-14.824 61.378Q-14.595 61.378-14.446 61.344Q-14.297 61.309-14.297 61.169L-14.297 59.529Q-14.297 59.194-14.417 58.994Q-14.537 58.794-14.851 58.794Q-15.121 58.794-15.355 58.930Q-15.589 59.067-15.728 59.301Q-15.866 59.535-15.866 59.809L-15.866 61.169Q-15.866 61.306-15.716 61.342Q-15.565 61.378-15.340 61.378\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M72.69 18.979h47.815\"\u002F>\u003Cpath stroke=\"none\" d=\"m123.105 18.979-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003Cg transform=\"translate(113.025 -46.412)\">\u003Cpath d=\"M-15.340 61.658L-16.974 61.658L-16.974 61.378Q-16.745 61.378-16.596 61.344Q-16.447 61.309-16.447 61.169L-16.447 57.550Q-16.447 57.280-16.555 57.218Q-16.663 57.157-16.974 57.157L-16.974 56.876L-15.894 56.801L-15.894 59.187Q-15.788 59.002-15.610 58.860Q-15.432 58.719-15.224 58.645Q-15.015 58.572-14.790 58.572Q-14.284 58.572-14 58.795Q-13.716 59.019-13.716 59.515L-13.716 61.169Q-13.716 61.306-13.568 61.342Q-13.419 61.378-13.193 61.378L-13.193 61.658L-14.824 61.658L-14.824 61.378Q-14.595 61.378-14.446 61.344Q-14.297 61.309-14.297 61.169L-14.297 59.529Q-14.297 59.194-14.417 58.994Q-14.537 58.794-14.851 58.794Q-15.121 58.794-15.355 58.930Q-15.589 59.067-15.728 59.301Q-15.866 59.535-15.866 59.809L-15.866 61.169Q-15.866 61.306-15.716 61.342Q-15.565 61.378-15.340 61.378\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M149.512 18.979h47.815\"\u002F>\u003Cpath stroke=\"none\" d=\"m199.927 18.979-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003Cg transform=\"translate(189.847 -46.412)\">\u003Cpath d=\"M-15.340 61.658L-16.974 61.658L-16.974 61.378Q-16.745 61.378-16.596 61.344Q-16.447 61.309-16.447 61.169L-16.447 57.550Q-16.447 57.280-16.555 57.218Q-16.663 57.157-16.974 57.157L-16.974 56.876L-15.894 56.801L-15.894 59.187Q-15.788 59.002-15.610 58.860Q-15.432 58.719-15.224 58.645Q-15.015 58.572-14.790 58.572Q-14.284 58.572-14 58.795Q-13.716 59.019-13.716 59.515L-13.716 61.169Q-13.716 61.306-13.568 61.342Q-13.419 61.378-13.193 61.378L-13.193 61.658L-14.824 61.658L-14.824 61.378Q-14.595 61.378-14.446 61.344Q-14.297 61.309-14.297 61.169L-14.297 59.529Q-14.297 59.194-14.417 58.994Q-14.537 58.794-14.851 58.794Q-15.121 58.794-15.355 58.930Q-15.589 59.067-15.728 59.301Q-15.866 59.535-15.866 59.809L-15.866 61.169Q-15.866 61.306-15.716 61.342Q-15.565 61.378-15.340 61.378\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg transform=\"translate(-44.934 1.674)\">\u003Cpath d=\"M-15.405 61.658L-16.957 61.658L-16.957 61.378Q-16.731 61.378-16.582 61.344Q-16.434 61.309-16.434 61.169L-16.434 59.320Q-16.434 59.132-16.482 59.048Q-16.529 58.965-16.627 58.946Q-16.724 58.927-16.936 58.927L-16.936 58.647L-15.880 58.572L-15.880 61.169Q-15.880 61.309-15.748 61.344Q-15.617 61.378-15.405 61.378L-15.405 61.658M-16.676 57.351Q-16.676 57.180-16.553 57.061Q-16.430 56.941-16.259 56.941Q-16.092 56.941-15.969 57.061Q-15.846 57.180-15.846 57.351Q-15.846 57.526-15.969 57.649Q-16.092 57.772-16.259 57.772Q-16.430 57.772-16.553 57.649Q-16.676 57.526-16.676 57.351M-13.077 61.658L-14.711 61.658L-14.711 61.378Q-14.482 61.378-14.333 61.344Q-14.185 61.309-14.185 61.169L-14.185 59.320Q-14.185 59.050-14.292 58.989Q-14.400 58.927-14.711 58.927L-14.711 58.647L-13.651 58.572L-13.651 59.221Q-13.481 58.913-13.176 58.742Q-12.872 58.572-12.527 58.572Q-12.021 58.572-11.737 58.795Q-11.454 59.019-11.454 59.515L-11.454 61.169Q-11.454 61.306-11.305 61.342Q-11.156 61.378-10.931 61.378L-10.931 61.658L-12.561 61.658L-12.561 61.378Q-12.332 61.378-12.183 61.344Q-12.035 61.309-12.035 61.169L-12.035 59.529Q-12.035 59.194-12.154 58.994Q-12.274 58.794-12.588 58.794Q-12.858 58.794-13.093 58.930Q-13.327 59.067-13.465 59.301Q-13.604 59.535-13.604 59.809L-13.604 61.169Q-13.604 61.306-13.453 61.342Q-13.303 61.378-13.077 61.378L-13.077 61.658M-8.699 63.015L-10.329 63.015L-10.329 62.735Q-10.100 62.735-9.951 62.700Q-9.803 62.666-9.803 62.526L-9.803 59.180Q-9.803 59.009-9.940 58.968Q-10.076 58.927-10.329 58.927L-10.329 58.647L-9.249 58.572L-9.249 58.978Q-9.027 58.777-8.740 58.674Q-8.453 58.572-8.145 58.572Q-7.718 58.572-7.354 58.785Q-6.990 58.999-6.776 59.363Q-6.563 59.727-6.563 60.147Q-6.563 60.592-6.802 60.956Q-7.041 61.320-7.434 61.523Q-7.827 61.726-8.272 61.726Q-8.538 61.726-8.786 61.626Q-9.034 61.525-9.222 61.344L-9.222 62.526Q-9.222 62.663-9.073 62.699Q-8.924 62.735-8.699 62.735L-8.699 63.015M-9.222 59.327L-9.222 60.937Q-9.088 61.190-8.846 61.347Q-8.603 61.504-8.326 61.504Q-7.998 61.504-7.745 61.303Q-7.492 61.101-7.359 60.783Q-7.226 60.465-7.226 60.147Q-7.226 59.918-7.291 59.689Q-7.356 59.460-7.484 59.262Q-7.612 59.064-7.807 58.944Q-8.002 58.825-8.234 58.825Q-8.528 58.825-8.796 58.954Q-9.065 59.084-9.222 59.327M-5.353 60.824L-5.353 59.320Q-5.353 59.050-5.460 58.989Q-5.568 58.927-5.879 58.927L-5.879 58.647L-4.772 58.572L-4.772 60.804L-4.772 60.824Q-4.772 61.104-4.720 61.248Q-4.669 61.391-4.527 61.448Q-4.385 61.504-4.098 61.504Q-3.845 61.504-3.640 61.364Q-3.435 61.224-3.319 60.998Q-3.203 60.773-3.203 60.523L-3.203 59.320Q-3.203 59.050-3.310 58.989Q-3.418 58.927-3.729 58.927L-3.729 58.647L-2.622 58.572L-2.622 60.985Q-2.622 61.176-2.569 61.258Q-2.516 61.340-2.415 61.359Q-2.314 61.378-2.099 61.378L-2.099 61.658L-3.175 61.726L-3.175 61.162Q-3.285 61.344-3.430 61.467Q-3.575 61.590-3.762 61.658Q-3.948 61.726-4.149 61.726Q-5.353 61.726-5.353 60.824M-0.984 60.817L-0.984 58.920L-1.624 58.920L-1.624 58.698Q-1.306 58.698-1.089 58.488Q-0.872 58.278-0.771 57.968Q-0.670 57.659-0.670 57.351L-0.403 57.351L-0.403 58.640L0.673 58.640L0.673 58.920L-0.403 58.920L-0.403 60.804Q-0.403 61.080-0.299 61.279Q-0.195 61.477 0.065 61.477Q0.222 61.477 0.328 61.373Q0.434 61.268 0.484 61.115Q0.533 60.961 0.533 60.804L0.533 60.390L0.800 60.390L0.800 60.817Q0.800 61.043 0.701 61.253Q0.602 61.463 0.417 61.595Q0.232 61.726 0.003 61.726Q-0.434 61.726-0.709 61.489Q-0.984 61.251-0.984 60.817\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(-43.906 -40.287)\">\u003Cpath d=\"M-14.837 61.658L-16.943 61.658L-16.943 61.378Q-16.222 61.378-16.222 61.169L-16.222 57.368Q-16.222 57.157-16.943 57.157L-16.943 56.876L-14.605 56.876Q-14.294 56.876-13.951 56.950Q-13.607 57.023-13.286 57.179Q-12.964 57.334-12.763 57.580Q-12.561 57.826-12.561 58.151Q-12.561 58.438-12.749 58.669Q-12.937 58.900-13.214 59.048Q-13.491 59.197-13.781 59.279Q-13.446 59.382-13.212 59.614Q-12.978 59.846-12.934 60.175L-12.848 60.783Q-12.821 60.995-12.775 61.164Q-12.729 61.333-12.624 61.453Q-12.520 61.573-12.332 61.573Q-12.117 61.573-12.001 61.388Q-11.884 61.203-11.884 60.971Q-11.867 60.906-11.792 60.889L-11.700 60.889Q-11.618 60.909-11.618 60.991Q-11.618 61.197-11.708 61.383Q-11.799 61.569-11.965 61.684Q-12.130 61.798-12.332 61.798Q-12.670 61.798-12.964 61.697Q-13.258 61.596-13.443 61.374Q-13.628 61.152-13.628 60.804L-13.628 60.195Q-13.628 59.946-13.771 59.759Q-13.915 59.573-14.138 59.474Q-14.362 59.375-14.612 59.375L-15.559 59.375L-15.559 61.169Q-15.559 61.378-14.837 61.378L-14.837 61.658M-15.559 57.368L-15.559 59.153L-14.704 59.153Q-14.417 59.153-14.178 59.110Q-13.939 59.067-13.751 58.958Q-13.563 58.848-13.451 58.650Q-13.340 58.452-13.340 58.151Q-13.340 57.574-13.708 57.365Q-14.075 57.157-14.704 57.157L-15.193 57.157Q-15.381 57.157-15.470 57.191Q-15.559 57.225-15.559 57.368M-9.420 61.658L-11.153 61.658L-11.153 61.378Q-10.432 61.378-10.432 60.978L-10.432 57.191Q-10.664 57.157-11.153 57.157L-11.153 56.876L-9.796 56.876Q-9.755 56.890-9.734 56.904L-7.031 60.551L-7.031 57.556Q-7.031 57.157-7.752 57.157L-7.752 56.876L-6.016 56.876L-6.016 57.157Q-6.737 57.157-6.737 57.556L-6.737 61.566Q-6.754 61.641-6.822 61.658L-6.942 61.658Q-6.990 61.648-7.003 61.624L-10.138 57.403L-10.138 60.978Q-10.138 61.378-9.420 61.378L-9.420 61.658M-3.521 61.658L-5.253 61.658L-5.253 61.378Q-4.532 61.378-4.532 60.978L-4.532 57.191Q-4.765 57.157-5.253 57.157L-5.253 56.876L-3.897 56.876Q-3.856 56.890-3.835 56.904L-1.131 60.551L-1.131 57.556Q-1.131 57.157-1.853 57.157L-1.853 56.876L-0.116 56.876L-0.116 57.157Q-0.837 57.157-0.837 57.556L-0.837 61.566Q-0.855 61.641-0.923 61.658L-1.043 61.658Q-1.090 61.648-1.104 61.624L-4.238 57.403L-4.238 60.978Q-4.238 61.378-3.521 61.378\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(-38.881 -122.761)\">\u003Cpath d=\"M-15.354 61.658L-16.957 61.658L-16.957 61.378Q-16.731 61.378-16.582 61.344Q-16.434 61.309-16.434 61.169L-16.434 57.550Q-16.434 57.280-16.541 57.218Q-16.649 57.157-16.957 57.157L-16.957 56.876L-15.880 56.801L-15.880 61.169Q-15.880 61.306-15.730 61.342Q-15.579 61.378-15.354 61.378L-15.354 61.658M-14.800 60.175Q-14.800 59.833-14.665 59.534Q-14.530 59.235-14.291 59.011Q-14.051 58.787-13.733 58.662Q-13.416 58.537-13.084 58.537Q-12.640 58.537-12.240 58.753Q-11.840 58.968-11.606 59.346Q-11.372 59.723-11.372 60.175Q-11.372 60.516-11.513 60.800Q-11.655 61.084-11.900 61.291Q-12.144 61.497-12.453 61.612Q-12.763 61.726-13.084 61.726Q-13.515 61.726-13.916 61.525Q-14.318 61.323-14.559 60.971Q-14.800 60.619-14.800 60.175M-13.084 61.477Q-12.482 61.477-12.259 61.099Q-12.035 60.721-12.035 60.089Q-12.035 59.477-12.269 59.118Q-12.503 58.760-13.084 58.760Q-14.137 58.760-14.137 60.089Q-14.137 60.721-13.911 61.099Q-13.686 61.477-13.084 61.477M-10.777 61.651L-10.777 60.588Q-10.777 60.564-10.750 60.537Q-10.722 60.510-10.698 60.510L-10.589 60.510Q-10.524 60.510-10.510 60.568Q-10.415 61.002-10.169 61.253Q-9.922 61.504-9.509 61.504Q-9.167 61.504-8.914 61.371Q-8.661 61.238-8.661 60.930Q-8.661 60.773-8.755 60.658Q-8.849 60.544-8.988 60.475Q-9.126 60.407-9.294 60.369L-9.875 60.270Q-10.230 60.202-10.503 59.981Q-10.777 59.761-10.777 59.419Q-10.777 59.170-10.666 58.995Q-10.555 58.821-10.368 58.722Q-10.182 58.623-9.967 58.580Q-9.752 58.537-9.509 58.537Q-9.095 58.537-8.815 58.719L-8.600 58.544Q-8.589 58.541-8.583 58.539Q-8.576 58.537-8.565 58.537L-8.514 58.537Q-8.487 58.537-8.463 58.561Q-8.439 58.585-8.439 58.613L-8.439 59.460Q-8.439 59.481-8.463 59.508Q-8.487 59.535-8.514 59.535L-8.627 59.535Q-8.654 59.535-8.680 59.510Q-8.706 59.484-8.706 59.460Q-8.706 59.224-8.812 59.060Q-8.918 58.896-9.100 58.814Q-9.283 58.732-9.516 58.732Q-9.844 58.732-10.100 58.835Q-10.357 58.937-10.357 59.214Q-10.357 59.409-10.174 59.518Q-9.991 59.628-9.762 59.669L-9.188 59.775Q-8.941 59.823-8.728 59.951Q-8.514 60.079-8.378 60.282Q-8.241 60.486-8.241 60.735Q-8.241 61.248-8.607 61.487Q-8.972 61.726-9.509 61.726Q-10.004 61.726-10.336 61.432L-10.603 61.706Q-10.623 61.726-10.650 61.726L-10.698 61.726Q-10.722 61.726-10.750 61.699Q-10.777 61.672-10.777 61.651M-7.612 61.651L-7.612 60.588Q-7.612 60.564-7.585 60.537Q-7.557 60.510-7.533 60.510L-7.424 60.510Q-7.359 60.510-7.345 60.568Q-7.250 61.002-7.003 61.253Q-6.757 61.504-6.344 61.504Q-6.002 61.504-5.749 61.371Q-5.496 61.238-5.496 60.930Q-5.496 60.773-5.590 60.658Q-5.684 60.544-5.823 60.475Q-5.961 60.407-6.128 60.369L-6.710 60.270Q-7.065 60.202-7.338 59.981Q-7.612 59.761-7.612 59.419Q-7.612 59.170-7.501 58.995Q-7.390 58.821-7.203 58.722Q-7.017 58.623-6.802 58.580Q-6.586 58.537-6.344 58.537Q-5.930 58.537-5.650 58.719L-5.435 58.544Q-5.424 58.541-5.418 58.539Q-5.411 58.537-5.400 58.537L-5.349 58.537Q-5.322 58.537-5.298 58.561Q-5.274 58.585-5.274 58.613L-5.274 59.460Q-5.274 59.481-5.298 59.508Q-5.322 59.535-5.349 59.535L-5.462 59.535Q-5.489 59.535-5.515 59.510Q-5.541 59.484-5.541 59.460Q-5.541 59.224-5.647 59.060Q-5.753 58.896-5.935 58.814Q-6.118 58.732-6.351 58.732Q-6.679 58.732-6.935 58.835Q-7.191 58.937-7.191 59.214Q-7.191 59.409-7.009 59.518Q-6.826 59.628-6.597 59.669L-6.023 59.775Q-5.776 59.823-5.563 59.951Q-5.349 60.079-5.212 60.282Q-5.076 60.486-5.076 60.735Q-5.076 61.248-5.441 61.487Q-5.807 61.726-6.344 61.726Q-6.839 61.726-7.171 61.432L-7.438 61.706Q-7.458 61.726-7.485 61.726L-7.533 61.726Q-7.557 61.726-7.585 61.699Q-7.612 61.672-7.612 61.651\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fsvg>\u003Cfigcaption class=\"tikz-cap\">Training an RNN language model. Each step reads its input embedding, updates the hidden state, and produces a softmax over the vocabulary; the loss is the negative log probability of the true next word, averaged across the sequence.\u003C\u002Ffigcaption>","\u003Csvg style=\"width:100%;max-width:400.434px;height:auto\" xmlns=\"http:\u002F\u002Fwww.w3.org\u002F2000\u002Fsvg\" viewBox=\"-75 -75 300.326 125.782\">\u003Cg stroke=\"currentColor\" style=\"stroke-miterlimit:10;stroke-width:.4\">\u003Cpath fill=\"none\" d=\"M-38.329 34.712h208.55\"\u002F>\u003Cpath stroke=\"none\" d=\"m172.222 34.712-3.2-1.6 1.2 1.6-1.2 1.6\"\u002F>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(214.083 1.75)\">\u003Cpath d=\"M-38.015 34.705L-38.015 33.642Q-38.015 33.618-37.987 33.591Q-37.960 33.564-37.936 33.564L-37.827 33.564Q-37.762 33.564-37.748 33.622Q-37.652 34.056-37.406 34.307Q-37.160 34.558-36.746 34.558Q-36.405 34.558-36.152 34.425Q-35.899 34.292-35.899 33.984Q-35.899 33.827-35.993 33.712Q-36.087 33.598-36.225 33.529Q-36.364 33.461-36.531 33.423L-37.112 33.324Q-37.468 33.256-37.741 33.035Q-38.015 32.815-38.015 32.473Q-38.015 32.224-37.903 32.049Q-37.792 31.875-37.606 31.776Q-37.420 31.677-37.204 31.634Q-36.989 31.591-36.746 31.591Q-36.333 31.591-36.053 31.773L-35.837 31.598Q-35.827 31.595-35.820 31.593Q-35.813 31.591-35.803 31.591L-35.752 31.591Q-35.725 31.591-35.701 31.615Q-35.677 31.639-35.677 31.667L-35.677 32.514Q-35.677 32.535-35.701 32.562Q-35.725 32.589-35.752 32.589L-35.865 32.589Q-35.892 32.589-35.918 32.564Q-35.943 32.538-35.943 32.514Q-35.943 32.278-36.049 32.114Q-36.155 31.950-36.338 31.868Q-36.521 31.786-36.753 31.786Q-37.081 31.786-37.338 31.889Q-37.594 31.991-37.594 32.268Q-37.594 32.463-37.411 32.572Q-37.228 32.682-36.999 32.723L-36.425 32.829Q-36.179 32.877-35.965 33.005Q-35.752 33.133-35.615 33.336Q-35.478 33.540-35.478 33.789Q-35.478 34.302-35.844 34.541Q-36.210 34.780-36.746 34.780Q-37.242 34.780-37.574 34.486L-37.840 34.760Q-37.861 34.780-37.888 34.780L-37.936 34.780Q-37.960 34.780-37.987 34.753Q-38.015 34.726-38.015 34.705M-34.323 33.871L-34.323 31.974L-34.962 31.974L-34.962 31.752Q-34.644 31.752-34.427 31.542Q-34.210 31.332-34.110 31.022Q-34.009 30.713-34.009 30.405L-33.742 30.405L-33.742 31.694L-32.665 31.694L-32.665 31.974L-33.742 31.974L-33.742 33.858Q-33.742 34.134-33.638 34.333Q-33.534 34.531-33.274 34.531Q-33.117 34.531-33.011 34.427Q-32.905 34.322-32.855 34.169Q-32.806 34.015-32.806 33.858L-32.806 33.444L-32.539 33.444L-32.539 33.871Q-32.539 34.097-32.638 34.307Q-32.737 34.517-32.922 34.649Q-33.106 34.780-33.335 34.780Q-33.773 34.780-34.048 34.543Q-34.323 34.305-34.323 33.871M-31.770 33.177Q-31.770 32.856-31.645 32.567Q-31.520 32.278-31.295 32.055Q-31.069 31.831-30.774 31.711Q-30.478 31.591-30.160 31.591Q-29.832 31.591-29.570 31.691Q-29.309 31.790-29.133 31.972Q-28.957 32.155-28.863 32.413Q-28.769 32.671-28.769 33.003Q-28.769 33.095-28.851 33.116L-31.107 33.116L-31.107 33.177Q-31.107 33.765-30.823 34.148Q-30.539 34.531-29.972 34.531Q-29.651 34.531-29.382 34.338Q-29.114 34.145-29.025 33.830Q-29.018 33.789-28.943 33.775L-28.851 33.775Q-28.769 33.799-28.769 33.871Q-28.769 33.878-28.776 33.905Q-28.889 34.302-29.259 34.541Q-29.630 34.780-30.054 34.780Q-30.492 34.780-30.892 34.572Q-31.291 34.363-31.531 33.996Q-31.770 33.629-31.770 33.177M-31.100 32.907L-29.285 32.907Q-29.285 32.630-29.382 32.378Q-29.480 32.125-29.678 31.969Q-29.876 31.814-30.160 31.814Q-30.437 31.814-30.651 31.972Q-30.864 32.131-30.982 32.386Q-31.100 32.641-31.100 32.907M-26.537 36.069L-28.167 36.069L-28.167 35.789Q-27.938 35.789-27.790 35.754Q-27.641 35.720-27.641 35.580L-27.641 32.234Q-27.641 32.063-27.778 32.022Q-27.914 31.981-28.167 31.981L-28.167 31.701L-27.087 31.626L-27.087 32.032Q-26.865 31.831-26.578 31.728Q-26.291 31.626-25.983 31.626Q-25.556 31.626-25.192 31.839Q-24.828 32.053-24.614 32.417Q-24.401 32.781-24.401 33.201Q-24.401 33.646-24.640 34.010Q-24.879 34.374-25.272 34.577Q-25.665 34.780-26.110 34.780Q-26.376 34.780-26.624 34.680Q-26.872 34.579-27.060 34.398L-27.060 35.580Q-27.060 35.717-26.911 35.753Q-26.763 35.789-26.537 35.789L-26.537 36.069M-27.060 32.381L-27.060 33.991Q-26.927 34.244-26.684 34.401Q-26.441 34.558-26.164 34.558Q-25.836 34.558-25.583 34.357Q-25.330 34.155-25.197 33.837Q-25.064 33.519-25.064 33.201Q-25.064 32.972-25.129 32.743Q-25.194 32.514-25.322 32.316Q-25.450 32.118-25.645 31.998Q-25.840 31.879-26.072 31.879Q-26.366 31.879-26.634 32.008Q-26.903 32.138-27.060 32.381M-23.765 34.705L-23.765 33.642Q-23.765 33.618-23.738 33.591Q-23.710 33.564-23.686 33.564L-23.577 33.564Q-23.512 33.564-23.498 33.622Q-23.403 34.056-23.157 34.307Q-22.911 34.558-22.497 34.558Q-22.155 34.558-21.902 34.425Q-21.649 34.292-21.649 33.984Q-21.649 33.827-21.743 33.712Q-21.837 33.598-21.976 33.529Q-22.114 33.461-22.282 33.423L-22.863 33.324Q-23.218 33.256-23.492 33.035Q-23.765 32.815-23.765 32.473Q-23.765 32.224-23.654 32.049Q-23.543 31.875-23.357 31.776Q-23.170 31.677-22.955 31.634Q-22.740 31.591-22.497 31.591Q-22.083 31.591-21.803 31.773L-21.588 31.598Q-21.578 31.595-21.571 31.593Q-21.564 31.591-21.554 31.591L-21.502 31.591Q-21.475 31.591-21.451 31.615Q-21.427 31.639-21.427 31.667L-21.427 32.514Q-21.427 32.535-21.451 32.562Q-21.475 32.589-21.502 32.589L-21.615 32.589Q-21.642 32.589-21.668 32.564Q-21.694 32.538-21.694 32.514Q-21.694 32.278-21.800 32.114Q-21.906 31.950-22.089 31.868Q-22.271 31.786-22.504 31.786Q-22.832 31.786-23.088 31.889Q-23.345 31.991-23.345 32.268Q-23.345 32.463-23.162 32.572Q-22.979 32.682-22.750 32.723L-22.176 32.829Q-21.930 32.877-21.716 33.005Q-21.502 33.133-21.366 33.336Q-21.229 33.540-21.229 33.789Q-21.229 34.302-21.595 34.541Q-21.960 34.780-22.497 34.780Q-22.993 34.780-23.324 34.486L-23.591 34.760Q-23.611 34.780-23.639 34.780L-23.686 34.780Q-23.710 34.780-23.738 34.753Q-23.765 34.726-23.765 34.705\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(214.083 1.75)\">\u003Cpath d=\"M-17.080 34.712L-17.347 34.712L-17.347 30.604Q-17.347 30.334-17.454 30.272Q-17.562 30.211-17.873 30.211L-17.873 29.930L-16.793 29.855L-16.793 32.025Q-16.584 31.834-16.299 31.730Q-16.014 31.626-15.716 31.626Q-15.398 31.626-15.101 31.747Q-14.804 31.868-14.581 32.084Q-14.359 32.299-14.233 32.584Q-14.106 32.870-14.106 33.201Q-14.106 33.646-14.346 34.010Q-14.585 34.374-14.978 34.577Q-15.371 34.780-15.815 34.780Q-16.010 34.780-16.200 34.724Q-16.389 34.668-16.550 34.563Q-16.711 34.459-16.851 34.298L-17.080 34.712M-16.765 32.367L-16.765 33.984Q-16.629 34.244-16.388 34.401Q-16.147 34.558-15.870 34.558Q-15.576 34.558-15.364 34.451Q-15.152 34.343-15.019 34.151Q-14.886 33.960-14.827 33.721Q-14.769 33.482-14.769 33.201Q-14.769 32.842-14.863 32.538Q-14.957 32.234-15.185 32.041Q-15.412 31.848-15.778 31.848Q-16.078 31.848-16.345 31.984Q-16.612 32.121-16.765 32.367M-13.412 33.984Q-13.412 33.652-13.189 33.425Q-12.965 33.198-12.621 33.070Q-12.278 32.941-11.905 32.889Q-11.533 32.836-11.228 32.836L-11.228 32.583Q-11.228 32.378-11.336 32.198Q-11.444 32.019-11.625 31.916Q-11.806 31.814-12.014 31.814Q-12.421 31.814-12.657 31.906Q-12.568 31.943-12.522 32.027Q-12.476 32.111-12.476 32.213Q-12.476 32.309-12.522 32.388Q-12.568 32.466-12.649 32.511Q-12.729 32.555-12.818 32.555Q-12.968 32.555-13.069 32.458Q-13.170 32.360-13.170 32.213Q-13.170 31.591-12.014 31.591Q-11.803 31.591-11.553 31.655Q-11.304 31.718-11.102 31.837Q-10.900 31.957-10.774 32.142Q-10.647 32.326-10.647 32.569L-10.647 34.145Q-10.647 34.261-10.586 34.357Q-10.524 34.452-10.411 34.452Q-10.302 34.452-10.237 34.358Q-10.172 34.264-10.172 34.145L-10.172 33.697L-9.906 33.697L-9.906 34.145Q-9.906 34.415-10.133 34.580Q-10.360 34.746-10.640 34.746Q-10.849 34.746-10.986 34.592Q-11.122 34.439-11.146 34.223Q-11.293 34.490-11.575 34.635Q-11.857 34.780-12.182 34.780Q-12.459 34.780-12.743 34.705Q-13.026 34.630-13.219 34.451Q-13.412 34.271-13.412 33.984M-12.797 33.984Q-12.797 34.158-12.696 34.288Q-12.596 34.418-12.440 34.488Q-12.284 34.558-12.120 34.558Q-11.902 34.558-11.693 34.461Q-11.485 34.363-11.357 34.182Q-11.228 34.001-11.228 33.775L-11.228 33.047Q-11.553 33.047-11.919 33.138Q-12.284 33.229-12.541 33.441Q-12.797 33.652-12.797 33.984M-9.489 33.201Q-9.489 32.873-9.354 32.572Q-9.219 32.272-8.983 32.051Q-8.747 31.831-8.443 31.711Q-8.139 31.591-7.814 31.591Q-7.308 31.591-6.959 31.694Q-6.611 31.796-6.611 32.172Q-6.611 32.319-6.708 32.420Q-6.805 32.521-6.952 32.521Q-7.106 32.521-7.205 32.422Q-7.305 32.323-7.305 32.172Q-7.305 31.984-7.164 31.892Q-7.366 31.841-7.807 31.841Q-8.162 31.841-8.391 32.037Q-8.620 32.234-8.721 32.543Q-8.822 32.853-8.822 33.201Q-8.822 33.550-8.696 33.856Q-8.569 34.162-8.315 34.346Q-8.060 34.531-7.704 34.531Q-7.482 34.531-7.298 34.447Q-7.113 34.363-6.978 34.208Q-6.843 34.052-6.785 33.844Q-6.771 33.789-6.717 33.789L-6.604 33.789Q-6.573 33.789-6.551 33.813Q-6.529 33.837-6.529 33.871L-6.529 33.892Q-6.614 34.179-6.802 34.377Q-6.990 34.575-7.255 34.678Q-7.520 34.780-7.814 34.780Q-8.244 34.780-8.632 34.574Q-9.020 34.367-9.254 34.004Q-9.489 33.642-9.489 33.201\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(214.083 1.75)\">\u003Cpath d=\"M-4.547 34.712L-6.130 34.712L-6.130 34.432Q-5.901 34.432-5.752 34.398Q-5.604 34.363-5.604 34.223L-5.604 30.604Q-5.604 30.334-5.711 30.272Q-5.819 30.211-6.130 30.211L-6.130 29.930L-5.050 29.855L-5.050 33.143L-4.065 32.374Q-3.860 32.237-3.860 32.087Q-3.860 32.043-3.901 32.008Q-3.942 31.974-3.987 31.974L-3.987 31.694L-2.623 31.694L-2.623 31.974Q-3.112 31.974-3.631 32.374L-4.188 32.808L-3.211 34.032Q-3.009 34.278-2.876 34.355Q-2.743 34.432-2.456 34.432L-2.456 34.712L-3.888 34.712L-3.888 34.432Q-3.700 34.432-3.700 34.319Q-3.700 34.223-3.854 34.032L-4.588 33.123L-5.070 33.502L-5.070 34.223Q-5.070 34.360-4.922 34.396Q-4.773 34.432-4.547 34.432\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(214.083 1.75)\">\u003Cpath d=\"M2.349 34.712L0.766 34.712L0.766 34.432Q0.995 34.432 1.144 34.398Q1.292 34.363 1.292 34.223L1.292 30.604Q1.292 30.334 1.185 30.272Q1.077 30.211 0.766 30.211L0.766 29.930L1.846 29.855L1.846 33.143L2.831 32.374Q3.036 32.237 3.036 32.087Q3.036 32.043 2.995 32.008Q2.954 31.974 2.909 31.974L2.909 31.694L4.273 31.694L4.273 31.974Q3.784 31.974 3.265 32.374L2.708 32.808L3.685 34.032Q3.887 34.278 4.020 34.355Q4.153 34.432 4.440 34.432L4.440 34.712L3.008 34.712L3.008 34.432Q3.196 34.432 3.196 34.319Q3.196 34.223 3.042 34.032L2.308 33.123L1.826 33.502L1.826 34.223Q1.826 34.360 1.974 34.396Q2.123 34.432 2.349 34.432\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-38.329 34.712v-91.894\"\u002F>\u003Cpath stroke=\"none\" d=\"m-38.329-59.182-1.6 3.2 1.6-1.2 1.6 1.2\"\u002F>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(-23.942 -98.788)\">\u003Cpath d=\"M-38.056 35.245Q-38.056 34.999-37.859 34.815Q-37.662 34.630-37.406 34.551Q-37.543 34.439-37.615 34.278Q-37.686 34.117-37.686 33.936Q-37.686 33.615-37.475 33.369Q-37.809 33.071-37.809 32.661Q-37.809 32.200-37.420 31.913Q-37.030 31.626-36.552 31.626Q-36.080 31.626-35.745 31.872Q-35.571 31.718-35.360 31.636Q-35.150 31.554-34.921 31.554Q-34.757 31.554-34.636 31.661Q-34.515 31.769-34.515 31.933Q-34.515 32.029-34.586 32.101Q-34.658 32.172-34.750 32.172Q-34.850 32.172-34.920 32.099Q-34.990 32.025-34.990 31.926Q-34.990 31.872-34.976 31.841L-34.969 31.827Q-34.962 31.807-34.954 31.796Q-34.945 31.786-34.942 31.779Q-35.297 31.779-35.584 32.002Q-35.297 32.295-35.297 32.661Q-35.297 32.976-35.482 33.208Q-35.666 33.441-35.955 33.569Q-36.244 33.697-36.552 33.697Q-36.753 33.697-36.945 33.647Q-37.136 33.598-37.314 33.488Q-37.406 33.615-37.406 33.758Q-37.406 33.940-37.278 34.075Q-37.150 34.210-36.965 34.210L-36.333 34.210Q-35.885 34.210-35.516 34.281Q-35.147 34.353-34.887 34.582Q-34.627 34.811-34.627 35.245Q-34.627 35.566-34.923 35.768Q-35.219 35.970-35.622 36.059Q-36.025 36.148-36.340 36.148Q-36.658 36.148-37.061 36.059Q-37.464 35.970-37.760 35.768Q-38.056 35.566-38.056 35.245M-37.601 35.245Q-37.601 35.474-37.382 35.623Q-37.163 35.772-36.871 35.840Q-36.579 35.908-36.340 35.908Q-36.176 35.908-35.967 35.872Q-35.759 35.837-35.552 35.756Q-35.345 35.676-35.214 35.548Q-35.082 35.420-35.082 35.245Q-35.082 34.893-35.463 34.799Q-35.844 34.705-36.347 34.705L-36.965 34.705Q-37.204 34.705-37.403 34.856Q-37.601 35.006-37.601 35.245M-36.552 33.458Q-35.885 33.458-35.885 32.661Q-35.885 31.861-36.552 31.861Q-37.222 31.861-37.222 32.661Q-37.222 33.458-36.552 33.458M-32.283 34.712L-34.019 34.712L-34.019 34.432Q-33.790 34.432-33.641 34.398Q-33.493 34.363-33.493 34.223L-33.493 32.374Q-33.493 32.104-33.600 32.043Q-33.708 31.981-34.019 31.981L-34.019 31.701L-32.990 31.626L-32.990 32.333Q-32.860 32.025-32.618 31.826Q-32.375 31.626-32.057 31.626Q-31.838 31.626-31.667 31.750Q-31.496 31.875-31.496 32.087Q-31.496 32.224-31.596 32.323Q-31.695 32.422-31.828 32.422Q-31.965 32.422-32.064 32.323Q-32.163 32.224-32.163 32.087Q-32.163 31.947-32.064 31.848Q-32.354 31.848-32.554 32.044Q-32.754 32.241-32.847 32.535Q-32.939 32.829-32.939 33.109L-32.939 34.223Q-32.939 34.432-32.283 34.432L-32.283 34.712M-30.854 33.984Q-30.854 33.652-30.630 33.425Q-30.406 33.198-30.063 33.070Q-29.719 32.941-29.347 32.889Q-28.974 32.836-28.670 32.836L-28.670 32.583Q-28.670 32.378-28.777 32.198Q-28.885 32.019-29.066 31.916Q-29.247 31.814-29.456 31.814Q-29.863 31.814-30.099 31.906Q-30.010 31.943-29.964 32.027Q-29.917 32.111-29.917 32.213Q-29.917 32.309-29.964 32.388Q-30.010 32.466-30.090 32.511Q-30.170 32.555-30.259 32.555Q-30.410 32.555-30.510 32.458Q-30.611 32.360-30.611 32.213Q-30.611 31.591-29.456 31.591Q-29.244 31.591-28.995 31.655Q-28.745 31.718-28.543 31.837Q-28.342 31.957-28.215 32.142Q-28.089 32.326-28.089 32.569L-28.089 34.145Q-28.089 34.261-28.027 34.357Q-27.966 34.452-27.853 34.452Q-27.744 34.452-27.679 34.358Q-27.614 34.264-27.614 34.145L-27.614 33.697L-27.347 33.697L-27.347 34.145Q-27.347 34.415-27.574 34.580Q-27.802 34.746-28.082 34.746Q-28.290 34.746-28.427 34.592Q-28.564 34.439-28.588 34.223Q-28.735 34.490-29.017 34.635Q-29.299 34.780-29.623 34.780Q-29.900 34.780-30.184 34.705Q-30.468 34.630-30.661 34.451Q-30.854 34.271-30.854 33.984M-30.239 33.984Q-30.239 34.158-30.138 34.288Q-30.037 34.418-29.881 34.488Q-29.726 34.558-29.562 34.558Q-29.343 34.558-29.135 34.461Q-28.926 34.363-28.798 34.182Q-28.670 34.001-28.670 33.775L-28.670 33.047Q-28.995 33.047-29.360 33.138Q-29.726 33.229-29.982 33.441Q-30.239 33.652-30.239 33.984M-26.930 33.201Q-26.930 32.863-26.790 32.572Q-26.650 32.282-26.405 32.068Q-26.161 31.855-25.857 31.740Q-25.553 31.626-25.228 31.626Q-24.958 31.626-24.695 31.725Q-24.432 31.824-24.240 32.002L-24.240 30.604Q-24.240 30.334-24.348 30.272Q-24.455 30.211-24.767 30.211L-24.767 29.930L-23.690 29.855L-23.690 34.039Q-23.690 34.227-23.635 34.310Q-23.580 34.394-23.480 34.413Q-23.379 34.432-23.163 34.432L-23.163 34.712L-24.271 34.780L-24.271 34.363Q-24.688 34.780-25.313 34.780Q-25.744 34.780-26.117 34.568Q-26.489 34.357-26.710 33.996Q-26.930 33.635-26.930 33.201M-25.255 34.558Q-25.047 34.558-24.860 34.486Q-24.674 34.415-24.520 34.278Q-24.367 34.141-24.271 33.963L-24.271 32.354Q-24.356 32.207-24.502 32.087Q-24.647 31.967-24.816 31.908Q-24.985 31.848-25.166 31.848Q-25.727 31.848-25.995 32.237Q-26.264 32.627-26.264 33.208Q-26.264 33.779-26.029 34.169Q-25.795 34.558-25.255 34.558M-20.897 34.712L-22.449 34.712L-22.449 34.432Q-22.224 34.432-22.075 34.398Q-21.926 34.363-21.926 34.223L-21.926 32.374Q-21.926 32.186-21.974 32.102Q-22.022 32.019-22.119 32Q-22.217 31.981-22.429 31.981L-22.429 31.701L-21.372 31.626L-21.372 34.223Q-21.372 34.363-21.241 34.398Q-21.109 34.432-20.897 34.432L-20.897 34.712M-22.169 30.405Q-22.169 30.234-22.046 30.115Q-21.923 29.995-21.752 29.995Q-21.584 29.995-21.461 30.115Q-21.338 30.234-21.338 30.405Q-21.338 30.580-21.461 30.703Q-21.584 30.826-21.752 30.826Q-21.923 30.826-22.046 30.703Q-22.169 30.580-22.169 30.405M-20.292 33.177Q-20.292 32.856-20.168 32.567Q-20.043 32.278-19.817 32.055Q-19.592 31.831-19.296 31.711Q-19 31.591-18.683 31.591Q-18.354 31.591-18.093 31.691Q-17.831 31.790-17.655 31.972Q-17.479 32.155-17.385 32.413Q-17.291 32.671-17.291 33.003Q-17.291 33.095-17.373 33.116L-19.629 33.116L-19.629 33.177Q-19.629 33.765-19.346 34.148Q-19.062 34.531-18.495 34.531Q-18.173 34.531-17.905 34.338Q-17.637 34.145-17.548 33.830Q-17.541 33.789-17.466 33.775L-17.373 33.775Q-17.291 33.799-17.291 33.871Q-17.291 33.878-17.298 33.905Q-17.411 34.302-17.782 34.541Q-18.153 34.780-18.577 34.780Q-19.014 34.780-19.414 34.572Q-19.814 34.363-20.053 33.996Q-20.292 33.629-20.292 33.177M-19.622 32.907L-17.808 32.907Q-17.808 32.630-17.905 32.378Q-18.002 32.125-18.201 31.969Q-18.399 31.814-18.683 31.814Q-18.959 31.814-19.173 31.972Q-19.387 32.131-19.505 32.386Q-19.622 32.641-19.622 32.907M-15.022 34.712L-16.656 34.712L-16.656 34.432Q-16.427 34.432-16.278 34.398Q-16.129 34.363-16.129 34.223L-16.129 32.374Q-16.129 32.104-16.237 32.043Q-16.345 31.981-16.656 31.981L-16.656 31.701L-15.596 31.626L-15.596 32.275Q-15.425 31.967-15.121 31.796Q-14.817 31.626-14.472 31.626Q-13.966 31.626-13.682 31.849Q-13.398 32.073-13.398 32.569L-13.398 34.223Q-13.398 34.360-13.250 34.396Q-13.101 34.432-12.875 34.432L-12.875 34.712L-14.506 34.712L-14.506 34.432Q-14.277 34.432-14.128 34.398Q-13.979 34.363-13.979 34.223L-13.979 32.583Q-13.979 32.248-14.099 32.048Q-14.219 31.848-14.533 31.848Q-14.803 31.848-15.037 31.984Q-15.271 32.121-15.410 32.355Q-15.548 32.589-15.548 32.863L-15.548 34.223Q-15.548 34.360-15.398 34.396Q-15.247 34.432-15.022 34.432\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(-23.942 -98.788)\">\u003Cpath d=\"M-11.953 33.871L-11.953 31.974L-12.592 31.974L-12.592 31.752Q-12.274 31.752-12.057 31.542Q-11.840 31.332-11.740 31.022Q-11.639 30.713-11.639 30.405L-11.372 30.405L-11.372 31.694L-10.295 31.694L-10.295 31.974L-11.372 31.974L-11.372 33.858Q-11.372 34.134-11.268 34.333Q-11.164 34.531-10.904 34.531Q-10.747 34.531-10.641 34.427Q-10.535 34.322-10.485 34.169Q-10.436 34.015-10.436 33.858L-10.436 33.444L-10.169 33.444L-10.169 33.871Q-10.169 34.097-10.268 34.307Q-10.367 34.517-10.552 34.649Q-10.736 34.780-10.965 34.780Q-11.403 34.780-11.678 34.543Q-11.953 34.305-11.953 33.871\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(-23.942 -98.788)\">\u003Cpath d=\"M-6.660 34.705L-6.660 33.642Q-6.660 33.618-6.632 33.591Q-6.605 33.564-6.581 33.564L-6.472 33.564Q-6.407 33.564-6.393 33.622Q-6.297 34.056-6.051 34.307Q-5.805 34.558-5.391 34.558Q-5.050 34.558-4.797 34.425Q-4.544 34.292-4.544 33.984Q-4.544 33.827-4.638 33.712Q-4.732 33.598-4.870 33.529Q-5.009 33.461-5.176 33.423L-5.757 33.324Q-6.113 33.256-6.386 33.035Q-6.660 32.815-6.660 32.473Q-6.660 32.224-6.548 32.049Q-6.437 31.875-6.251 31.776Q-6.065 31.677-5.849 31.634Q-5.634 31.591-5.391 31.591Q-4.978 31.591-4.698 31.773L-4.482 31.598Q-4.472 31.595-4.465 31.593Q-4.458 31.591-4.448 31.591L-4.397 31.591Q-4.370 31.591-4.346 31.615Q-4.322 31.639-4.322 31.667L-4.322 32.514Q-4.322 32.535-4.346 32.562Q-4.370 32.589-4.397 32.589L-4.510 32.589Q-4.537 32.589-4.563 32.564Q-4.588 32.538-4.588 32.514Q-4.588 32.278-4.694 32.114Q-4.800 31.950-4.983 31.868Q-5.166 31.786-5.398 31.786Q-5.726 31.786-5.983 31.889Q-6.239 31.991-6.239 32.268Q-6.239 32.463-6.056 32.572Q-5.873 32.682-5.644 32.723L-5.070 32.829Q-4.824 32.877-4.610 33.005Q-4.397 33.133-4.260 33.336Q-4.123 33.540-4.123 33.789Q-4.123 34.302-4.489 34.541Q-4.855 34.780-5.391 34.780Q-5.887 34.780-6.219 34.486L-6.485 34.760Q-6.506 34.780-6.533 34.780L-6.581 34.780Q-6.605 34.780-6.632 34.753Q-6.660 34.726-6.660 34.705M-3.495 33.201Q-3.495 32.873-3.359 32.572Q-3.224 32.272-2.989 32.051Q-2.753 31.831-2.449 31.711Q-2.144 31.591-1.820 31.591Q-1.314 31.591-0.965 31.694Q-0.617 31.796-0.617 32.172Q-0.617 32.319-0.714 32.420Q-0.811 32.521-0.958 32.521Q-1.112 32.521-1.211 32.422Q-1.310 32.323-1.310 32.172Q-1.310 31.984-1.170 31.892Q-1.372 31.841-1.813 31.841Q-2.168 31.841-2.397 32.037Q-2.626 32.234-2.727 32.543Q-2.828 32.853-2.828 33.201Q-2.828 33.550-2.702 33.856Q-2.575 34.162-2.320 34.346Q-2.066 34.531-1.710 34.531Q-1.488 34.531-1.304 34.447Q-1.119 34.363-0.984 34.208Q-0.849 34.052-0.791 33.844Q-0.777 33.789-0.723 33.789L-0.610 33.789Q-0.579 33.789-0.557 33.813Q-0.535 33.837-0.535 33.871L-0.535 33.892Q-0.620 34.179-0.808 34.377Q-0.996 34.575-1.261 34.678Q-1.526 34.780-1.820 34.780Q-2.250 34.780-2.638 34.574Q-3.026 34.367-3.260 34.004Q-3.495 33.642-3.495 33.201M0.111 33.984Q0.111 33.652 0.335 33.425Q0.559 33.198 0.903 33.070Q1.246 32.941 1.619 32.889Q1.991 32.836 2.296 32.836L2.296 32.583Q2.296 32.378 2.188 32.198Q2.080 32.019 1.899 31.916Q1.718 31.814 1.509 31.814Q1.103 31.814 0.867 31.906Q0.956 31.943 1.002 32.027Q1.048 32.111 1.048 32.213Q1.048 32.309 1.002 32.388Q0.956 32.466 0.875 32.511Q0.795 32.555 0.706 32.555Q0.556 32.555 0.455 32.458Q0.354 32.360 0.354 32.213Q0.354 31.591 1.509 31.591Q1.721 31.591 1.971 31.655Q2.220 31.718 2.422 31.837Q2.624 31.957 2.750 32.142Q2.877 32.326 2.877 32.569L2.877 34.145Q2.877 34.261 2.938 34.357Q3 34.452 3.112 34.452Q3.222 34.452 3.287 34.358Q3.352 34.264 3.352 34.145L3.352 33.697L3.618 33.697L3.618 34.145Q3.618 34.415 3.391 34.580Q3.164 34.746 2.883 34.746Q2.675 34.746 2.538 34.592Q2.401 34.439 2.378 34.223Q2.231 34.490 1.949 34.635Q1.667 34.780 1.342 34.780Q1.065 34.780 0.781 34.705Q0.498 34.630 0.305 34.451Q0.111 34.271 0.111 33.984M0.727 33.984Q0.727 34.158 0.828 34.288Q0.928 34.418 1.084 34.488Q1.239 34.558 1.403 34.558Q1.622 34.558 1.831 34.461Q2.039 34.363 2.167 34.182Q2.296 34.001 2.296 33.775L2.296 33.047Q1.971 33.047 1.605 33.138Q1.239 33.229 0.983 33.441Q0.727 33.652 0.727 33.984M5.703 34.712L4.100 34.712L4.100 34.432Q4.326 34.432 4.474 34.398Q4.623 34.363 4.623 34.223L4.623 30.604Q4.623 30.334 4.516 30.272Q4.408 30.211 4.100 30.211L4.100 29.930L5.177 29.855L5.177 34.223Q5.177 34.360 5.327 34.396Q5.478 34.432 5.703 34.432L5.703 34.712M6.257 33.177Q6.257 32.856 6.382 32.567Q6.506 32.278 6.732 32.055Q6.958 31.831 7.253 31.711Q7.549 31.591 7.867 31.591Q8.195 31.591 8.456 31.691Q8.718 31.790 8.894 31.972Q9.070 32.155 9.164 32.413Q9.258 32.671 9.258 33.003Q9.258 33.095 9.176 33.116L6.920 33.116L6.920 33.177Q6.920 33.765 7.204 34.148Q7.487 34.531 8.055 34.531Q8.376 34.531 8.644 34.338Q8.913 34.145 9.002 33.830Q9.008 33.789 9.084 33.775L9.176 33.775Q9.258 33.799 9.258 33.871Q9.258 33.878 9.251 33.905Q9.138 34.302 8.767 34.541Q8.397 34.780 7.973 34.780Q7.535 34.780 7.135 34.572Q6.735 34.363 6.496 33.996Q6.257 33.629 6.257 33.177M6.927 32.907L8.742 32.907Q8.742 32.630 8.644 32.378Q8.547 32.125 8.349 31.969Q8.151 31.814 7.867 31.814Q7.590 31.814 7.376 31.972Q7.163 32.131 7.045 32.386Q6.927 32.641 6.927 32.907\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" stroke=\"var(--tk-accent)\" d=\"m-38.329-50.646 4.961 10.464 4.961 9.195 4.96 8.06 4.962 7.079 4.96 6.206 4.962 5.447L-3.603.58 1.359 4.77 6.32 8.45l4.961 3.22 4.961 2.833 4.961 2.476 4.96 2.178 4.962 1.912 4.96 1.674 4.962 1.469 4.96 1.288 4.961 1.136 4.961.99 4.961.87 4.961.762 4.961.674 4.96.589 4.962.512 4.96.453 4.962.401 4.96.341 4.961.308 4.962.264 4.96.24 4.961.204 4.961.18 4.961.162 4.961.136 4.96.12 4.962.106 4.96.094 4.962.083 4.96.072\" style=\"stroke-width:1.2\"\u002F>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\">\u003Cg fill=\"var(--tk-accent)\" stroke=\"none\" font-family=\"cmr8\" font-size=\"8\">\u003Cg transform=\"translate(145.797 -12.871)\">\u003Cpath d=\"M-36.208 34.880Q-36.911 34.880-37.311 34.480Q-37.712 34.079-37.856 33.470Q-38.001 32.860-38.001 32.161Q-38.001 31.638-37.931 31.175Q-37.860 30.712-37.667 30.300Q-37.474 29.888-37.116 29.640Q-36.759 29.392-36.208 29.392Q-35.657 29.392-35.300 29.640Q-34.942 29.888-34.751 30.298Q-34.559 30.708-34.489 31.177Q-34.419 31.646-34.419 32.161Q-34.419 32.860-34.561 33.468Q-34.704 34.075-35.104 34.478Q-35.505 34.880-36.208 34.880M-36.208 34.622Q-35.735 34.622-35.503 34.187Q-35.270 33.751-35.216 33.212Q-35.161 32.673-35.161 32.032Q-35.161 31.036-35.345 30.343Q-35.528 29.650-36.208 29.650Q-36.575 29.650-36.796 29.888Q-37.017 30.126-37.112 30.483Q-37.208 30.841-37.233 31.212Q-37.259 31.583-37.259 32.032Q-37.259 32.673-37.204 33.212Q-37.149 33.751-36.917 34.187Q-36.684 34.622-36.208 34.622M-33.364 34.247Q-33.364 34.064-33.227 33.927Q-33.091 33.790-32.899 33.790Q-32.708 33.790-32.575 33.923Q-32.442 34.056-32.442 34.247Q-32.442 34.446-32.575 34.579Q-32.708 34.712-32.899 34.712Q-33.091 34.712-33.227 34.575Q-33.364 34.439-33.364 34.247M-29.602 34.880Q-30.274 34.880-30.671 34.456Q-31.067 34.032-31.220 33.413Q-31.372 32.794-31.372 32.126Q-31.372 31.466-31.100 30.833Q-30.829 30.200-30.315 29.796Q-29.802 29.392-29.130 29.392Q-28.841 29.392-28.593 29.491Q-28.345 29.591-28.198 29.792Q-28.052 29.993-28.052 30.298Q-28.052 30.403-28.102 30.495Q-28.153 30.587-28.245 30.638Q-28.337 30.689-28.442 30.689Q-28.610 30.689-28.724 30.575Q-28.837 30.462-28.837 30.298Q-28.837 30.138-28.727 30.021Q-28.618 29.903-28.450 29.903Q-28.649 29.634-29.130 29.634Q-29.548 29.634-29.880 29.911Q-30.212 30.189-30.388 30.607Q-30.587 31.107-30.587 32.009Q-30.423 31.685-30.143 31.485Q-29.864 31.286-29.517 31.286Q-29.032 31.286-28.647 31.532Q-28.263 31.778-28.050 32.187Q-27.837 32.595-27.837 33.079Q-27.837 33.571-28.067 33.983Q-28.298 34.396-28.708 34.638Q-29.118 34.880-29.602 34.880M-29.602 34.607Q-29.177 34.607-28.960 34.386Q-28.743 34.165-28.681 33.839Q-28.618 33.513-28.618 33.079Q-28.618 32.767-28.643 32.517Q-28.669 32.267-28.759 32.042Q-28.849 31.817-29.044 31.681Q-29.239 31.544-29.556 31.544Q-29.884 31.544-30.116 31.753Q-30.349 31.962-30.460 32.280Q-30.571 32.599-30.571 32.911Q-30.567 32.950-30.565 32.983Q-30.563 33.017-30.563 33.071Q-30.563 33.087-30.565 33.095Q-30.567 33.103-30.571 33.110Q-30.571 33.685-30.345 34.146Q-30.118 34.607-29.602 34.607\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.240\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(145.797 -12.871)\">\u003Cpath d=\"M-23.771 33.751L-23.771 31.560L-24.474 31.560L-24.474 31.306Q-24.118 31.306-23.876 31.073Q-23.634 30.841-23.523 30.493Q-23.411 30.146-23.411 29.790L-23.130 29.790L-23.130 31.263L-21.954 31.263L-21.954 31.560L-23.130 31.560L-23.130 33.735Q-23.130 34.056-23.011 34.284Q-22.892 34.513-22.611 34.513Q-22.431 34.513-22.314 34.390Q-22.197 34.267-22.144 34.087Q-22.091 33.907-22.091 33.735L-22.091 33.263L-21.810 33.263L-21.810 33.751Q-21.810 34.005-21.915 34.245Q-22.021 34.485-22.218 34.638Q-22.415 34.790-22.673 34.790Q-22.989 34.790-23.241 34.667Q-23.493 34.544-23.632 34.310Q-23.771 34.075-23.771 33.751M-21.091 33.017Q-21.091 32.513-20.835 32.081Q-20.579 31.650-20.144 31.398Q-19.708 31.146-19.208 31.146Q-18.822 31.146-18.480 31.290Q-18.138 31.435-17.876 31.696Q-17.614 31.958-17.472 32.294Q-17.329 32.630-17.329 33.017Q-17.329 33.509-17.593 33.919Q-17.857 34.329-18.286 34.560Q-18.716 34.790-19.208 34.790Q-19.700 34.790-20.134 34.558Q-20.568 34.325-20.829 33.917Q-21.091 33.509-21.091 33.017M-19.208 34.513Q-18.751 34.513-18.499 34.290Q-18.247 34.067-18.159 33.716Q-18.072 33.364-18.072 32.919Q-18.072 32.489-18.165 32.151Q-18.259 31.814-18.513 31.607Q-18.767 31.400-19.208 31.400Q-19.857 31.400-20.101 31.816Q-20.345 32.232-20.345 32.919Q-20.345 33.364-20.257 33.716Q-20.169 34.067-19.917 34.290Q-19.665 34.513-19.208 34.513\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.240\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(145.797 -12.871)\">\u003Cpath d=\"M-13.382 33.751L-13.382 31.560L-14.085 31.560L-14.085 31.306Q-13.729 31.306-13.487 31.073Q-13.245 30.841-13.134 30.493Q-13.022 30.146-13.022 29.790L-12.741 29.790L-12.741 31.263L-11.565 31.263L-11.565 31.560L-12.741 31.560L-12.741 33.735Q-12.741 34.056-12.622 34.284Q-12.503 34.513-12.222 34.513Q-12.042 34.513-11.925 34.390Q-11.807 34.267-11.755 34.087Q-11.702 33.907-11.702 33.735L-11.702 33.263L-11.421 33.263L-11.421 33.751Q-11.421 34.005-11.526 34.245Q-11.632 34.485-11.829 34.638Q-12.026 34.790-12.284 34.790Q-12.600 34.790-12.852 34.667Q-13.104 34.544-13.243 34.310Q-13.382 34.075-13.382 33.751M-8.772 34.712L-10.628 34.712L-10.628 34.415Q-10.354 34.415-10.186 34.368Q-10.018 34.321-10.018 34.153L-10.018 29.993Q-10.018 29.778-10.081 29.683Q-10.143 29.587-10.263 29.566Q-10.382 29.544-10.628 29.544L-10.628 29.247L-9.405 29.161L-9.405 31.864Q-9.280 31.653-9.093 31.503Q-8.905 31.353-8.679 31.269Q-8.452 31.185-8.206 31.185Q-7.038 31.185-7.038 32.263L-7.038 34.153Q-7.038 34.321-6.868 34.368Q-6.698 34.415-6.429 34.415L-6.429 34.712L-8.284 34.712L-8.284 34.415Q-8.011 34.415-7.843 34.368Q-7.675 34.321-7.675 34.153L-7.675 32.278Q-7.675 31.896-7.796 31.667Q-7.917 31.439-8.268 31.439Q-8.581 31.439-8.835 31.601Q-9.089 31.763-9.235 32.032Q-9.382 32.302-9.382 32.599L-9.382 34.153Q-9.382 34.321-9.212 34.368Q-9.042 34.415-8.772 34.415L-8.772 34.712M-5.983 32.958Q-5.983 32.478-5.751 32.062Q-5.518 31.646-5.108 31.396Q-4.698 31.146-4.222 31.146Q-3.491 31.146-3.093 31.587Q-2.694 32.028-2.694 32.759Q-2.694 32.864-2.788 32.888L-5.237 32.888L-5.237 32.958Q-5.237 33.368-5.116 33.724Q-4.995 34.079-4.724 34.296Q-4.452 34.513-4.022 34.513Q-3.659 34.513-3.362 34.284Q-3.065 34.056-2.964 33.704Q-2.956 33.657-2.870 33.642L-2.788 33.642Q-2.694 33.669-2.694 33.751Q-2.694 33.759-2.702 33.790Q-2.765 34.017-2.903 34.200Q-3.042 34.384-3.233 34.517Q-3.425 34.650-3.643 34.720Q-3.862 34.790-4.100 34.790Q-4.472 34.790-4.809 34.653Q-5.147 34.517-5.415 34.265Q-5.682 34.013-5.833 33.673Q-5.983 33.333-5.983 32.958M-5.229 32.650L-3.268 32.650Q-3.268 32.345-3.370 32.054Q-3.472 31.763-3.688 31.581Q-3.905 31.400-4.222 31.400Q-4.522 31.400-4.753 31.587Q-4.983 31.775-5.106 32.066Q-5.229 32.357-5.229 32.650\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.240\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(145.797 -12.871)\">\u003Cpath d=\"M2.457 34.712L0.660 34.712L0.660 34.415Q0.929 34.415 1.097 34.370Q1.265 34.325 1.265 34.153L1.265 29.993Q1.265 29.778 1.203 29.683Q1.140 29.587 1.023 29.566Q0.906 29.544 0.660 29.544L0.660 29.247L1.882 29.161L1.882 32.927L2.980 32.040Q3.187 31.860 3.187 31.712Q3.187 31.646 3.134 31.603Q3.082 31.560 3.011 31.560L3.011 31.263L4.546 31.263L4.546 31.560Q4.015 31.560 3.417 32.040L2.808 32.536L3.882 33.935Q4.019 34.110 4.126 34.218Q4.234 34.325 4.369 34.370Q4.503 34.415 4.730 34.415L4.730 34.712L3.105 34.712L3.105 34.415Q3.347 34.415 3.347 34.263Q3.347 34.185 3.304 34.114Q3.261 34.044 3.179 33.935L2.378 32.888L1.851 33.314L1.851 34.153Q1.851 34.321 2.019 34.368Q2.187 34.415 2.457 34.415\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.240\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003Cg transform=\"translate(-1.993 9.467)\">\u003Cpath d=\"M-35.003 34.712L-37.533 34.712L-37.533 34.432Q-36.565 34.432-36.565 34.223L-36.565 30.604Q-36.958 30.792-37.580 30.792L-37.580 30.511Q-37.163 30.511-36.799 30.410Q-36.435 30.310-36.179 30.064L-36.053 30.064Q-35.988 30.081-35.971 30.149L-35.971 34.223Q-35.971 34.432-35.003 34.432\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(94.745 9.467)\">\u003Cpath d=\"M-37.522 33.950L-37.553 33.950Q-37.416 34.247-37.119 34.423Q-36.822 34.599-36.494 34.599Q-36.131 34.599-35.904 34.421Q-35.677 34.244-35.583 33.955Q-35.489 33.666-35.489 33.304Q-35.489 32.989-35.543 32.704Q-35.598 32.419-35.771 32.213Q-35.943 32.008-36.258 32.008Q-36.531 32.008-36.714 32.075Q-36.897 32.142-37.001 32.231Q-37.105 32.319-37.201 32.429Q-37.297 32.538-37.341 32.548L-37.420 32.548Q-37.492 32.531-37.509 32.460L-37.509 30.142Q-37.509 30.108-37.485 30.086Q-37.461 30.064-37.427 30.064L-37.399 30.064Q-37.112 30.180-36.844 30.234Q-36.576 30.289-36.299 30.289Q-36.022 30.289-35.752 30.234Q-35.482 30.180-35.202 30.064L-35.178 30.064Q-35.143 30.064-35.120 30.087Q-35.096 30.111-35.096 30.142L-35.096 30.211Q-35.096 30.238-35.116 30.258Q-35.390 30.573-35.774 30.749Q-36.159 30.925-36.572 30.925Q-36.911 30.925-37.228 30.839L-37.228 32.121Q-36.832 31.786-36.258 31.786Q-35.854 31.786-35.518 31.996Q-35.181 32.207-34.988 32.559Q-34.795 32.911-34.795 33.311Q-34.795 33.642-34.935 33.928Q-35.075 34.213-35.319 34.423Q-35.564 34.633-35.866 34.743Q-36.169 34.852-36.487 34.852Q-36.846 34.852-37.172 34.688Q-37.498 34.524-37.693 34.232Q-37.888 33.940-37.888 33.577Q-37.888 33.427-37.782 33.321Q-37.676 33.215-37.522 33.215Q-37.369 33.215-37.264 33.319Q-37.160 33.423-37.160 33.577Q-37.160 33.734-37.264 33.842Q-37.369 33.950-37.522 33.950\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(189.49 9.467)\">\u003Cpath d=\"M-35.003 34.712L-37.533 34.712L-37.533 34.432Q-36.565 34.432-36.565 34.223L-36.565 30.604Q-36.958 30.792-37.580 30.792L-37.580 30.511Q-37.163 30.511-36.799 30.410Q-36.435 30.310-36.179 30.064L-36.053 30.064Q-35.988 30.081-35.971 30.149L-35.971 34.223Q-35.971 34.432-35.003 34.432L-35.003 34.712M-32.358 34.852Q-32.994 34.852-33.358 34.507Q-33.722 34.162-33.857 33.637Q-33.992 33.112-33.992 32.487Q-33.992 31.462-33.636 30.763Q-33.281 30.064-32.358 30.064Q-31.432 30.064-31.079 30.763Q-30.727 31.462-30.727 32.487Q-30.727 33.112-30.862 33.637Q-30.997 34.162-31.360 34.507Q-31.722 34.852-32.358 34.852M-32.358 34.627Q-31.920 34.627-31.707 34.252Q-31.493 33.878-31.444 33.411Q-31.394 32.945-31.394 32.367Q-31.394 31.814-31.444 31.386Q-31.493 30.959-31.705 30.624Q-31.917 30.289-32.358 30.289Q-32.700 30.289-32.903 30.496Q-33.106 30.703-33.194 31.015Q-33.281 31.328-33.303 31.644Q-33.325 31.961-33.325 32.367Q-33.325 32.784-33.303 33.126Q-33.281 33.468-33.192 33.816Q-33.103 34.165-32.898 34.396Q-32.693 34.627-32.358 34.627\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" stroke=\"var(--tk-warn)\" d=\"M-38.329 32.152h199.17\" style=\"stroke-dasharray:3.0,3.0\"\u002F>\u003Cg fill=\"var(--tk-warn)\" stroke=\"var(--tk-warn)\">\u003Cg fill=\"var(--tk-warn)\" stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(158.789 -6.378)\">\u003Cpath d=\"M-37.208 34.712L-37.475 34.712L-37.475 30.604Q-37.475 30.334-37.582 30.272Q-37.690 30.211-38.001 30.211L-38.001 29.930L-36.921 29.855L-36.921 32.025Q-36.712 31.834-36.427 31.730Q-36.142 31.626-35.844 31.626Q-35.526 31.626-35.229 31.747Q-34.932 31.868-34.709 32.084Q-34.487 32.299-34.361 32.584Q-34.234 32.870-34.234 33.201Q-34.234 33.646-34.474 34.010Q-34.713 34.374-35.106 34.577Q-35.499 34.780-35.943 34.780Q-36.138 34.780-36.328 34.724Q-36.517 34.668-36.678 34.563Q-36.839 34.459-36.979 34.298L-37.208 34.712M-36.893 32.367L-36.893 33.984Q-36.757 34.244-36.516 34.401Q-36.275 34.558-35.998 34.558Q-35.704 34.558-35.492 34.451Q-35.280 34.343-35.147 34.151Q-35.014 33.960-34.955 33.721Q-34.897 33.482-34.897 33.201Q-34.897 32.842-34.991 32.538Q-35.085 32.234-35.313 32.041Q-35.540 31.848-35.906 31.848Q-36.206 31.848-36.473 31.984Q-36.740 32.121-36.893 32.367\" fill=\"var(--tk-warn)\" stroke=\"var(--tk-warn)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(158.789 -6.378)\">\u003Cpath d=\"M-33.424 33.177Q-33.424 32.856-33.299 32.567Q-33.174 32.278-32.948 32.055Q-32.723 31.831-32.427 31.711Q-32.132 31.591-31.814 31.591Q-31.486 31.591-31.224 31.691Q-30.963 31.790-30.787 31.972Q-30.611 32.155-30.517 32.413Q-30.423 32.671-30.423 33.003Q-30.423 33.095-30.505 33.116L-32.760 33.116L-32.760 33.177Q-32.760 33.765-32.477 34.148Q-32.193 34.531-31.626 34.531Q-31.304 34.531-31.036 34.338Q-30.768 34.145-30.679 33.830Q-30.672 33.789-30.597 33.775L-30.505 33.775Q-30.423 33.799-30.423 33.871Q-30.423 33.878-30.429 33.905Q-30.542 34.302-30.913 34.541Q-31.284 34.780-31.708 34.780Q-32.145 34.780-32.545 34.572Q-32.945 34.363-33.184 33.996Q-33.424 33.629-33.424 33.177M-32.754 32.907L-30.939 32.907Q-30.939 32.630-31.036 32.378Q-31.134 32.125-31.332 31.969Q-31.530 31.814-31.814 31.814Q-32.091 31.814-32.304 31.972Q-32.518 32.131-32.636 32.386Q-32.754 32.641-32.754 32.907M-28.167 34.712L-29.770 34.712L-29.770 34.432Q-29.544 34.432-29.395 34.398Q-29.247 34.363-29.247 34.223L-29.247 30.604Q-29.247 30.334-29.354 30.272Q-29.462 30.211-29.770 30.211L-29.770 29.930L-28.693 29.855L-28.693 34.223Q-28.693 34.360-28.543 34.396Q-28.392 34.432-28.167 34.432L-28.167 34.712M-27.613 33.229Q-27.613 32.887-27.478 32.588Q-27.343 32.289-27.104 32.065Q-26.864 31.841-26.547 31.716Q-26.229 31.591-25.897 31.591Q-25.453 31.591-25.053 31.807Q-24.653 32.022-24.419 32.400Q-24.185 32.777-24.185 33.229Q-24.185 33.570-24.327 33.854Q-24.468 34.138-24.713 34.345Q-24.957 34.551-25.267 34.666Q-25.576 34.780-25.897 34.780Q-26.328 34.780-26.729 34.579Q-27.131 34.377-27.372 34.025Q-27.613 33.673-27.613 33.229M-25.897 34.531Q-25.296 34.531-25.072 34.153Q-24.848 33.775-24.848 33.143Q-24.848 32.531-25.082 32.172Q-25.316 31.814-25.897 31.814Q-26.950 31.814-26.950 33.143Q-26.950 33.775-26.724 34.153Q-26.499 34.531-25.897 34.531\" fill=\"var(--tk-warn)\" stroke=\"var(--tk-warn)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(158.789 -6.378)\">\u003Cpath d=\"M-22.404 34.685L-23.385 32.186Q-23.446 32.043-23.564 32.008Q-23.682 31.974-23.898 31.974L-23.898 31.694L-22.418 31.694L-22.418 31.974Q-22.797 31.974-22.797 32.135Q-22.797 32.145-22.783 32.186L-22.069 34.018L-21.396 32.313Q-21.426 32.241-21.426 32.213Q-21.426 32.186-21.454 32.186Q-21.515 32.039-21.633 32.007Q-21.751 31.974-21.963 31.974L-21.963 31.694L-20.565 31.694L-20.565 31.974Q-20.941 31.974-20.941 32.135Q-20.941 32.166-20.934 32.186L-20.179 34.124L-19.492 32.374Q-19.471 32.323-19.471 32.268Q-19.471 32.128-19.584 32.051Q-19.697 31.974-19.837 31.974L-19.837 31.694L-18.617 31.694L-18.617 31.974Q-18.822 31.974-18.977 32.080Q-19.133 32.186-19.205 32.374L-20.110 34.685Q-20.145 34.780-20.257 34.780L-20.326 34.780Q-20.435 34.780-20.473 34.685L-21.255 32.682L-22.042 34.685Q-22.076 34.780-22.189 34.780L-22.257 34.780Q-22.366 34.780-22.404 34.685\" fill=\"var(--tk-warn)\" stroke=\"var(--tk-warn)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(158.789 -6.378)\">\u003Cpath d=\"M-13.714 34.852Q-14.349 34.852-14.713 34.507Q-15.078 34.162-15.213 33.637Q-15.348 33.112-15.348 32.487Q-15.348 31.462-14.992 30.763Q-14.637 30.064-13.714 30.064Q-12.787 30.064-12.435 30.763Q-12.083 31.462-12.083 32.487Q-12.083 33.112-12.218 33.637Q-12.353 34.162-12.716 34.507Q-13.078 34.852-13.714 34.852M-13.714 34.627Q-13.276 34.627-13.063 34.252Q-12.849 33.878-12.799 33.411Q-12.750 32.945-12.750 32.367Q-12.750 31.814-12.799 31.386Q-12.849 30.959-13.061 30.624Q-13.273 30.289-13.714 30.289Q-14.056 30.289-14.259 30.496Q-14.462 30.703-14.549 31.015Q-14.637 31.328-14.659 31.644Q-14.681 31.961-14.681 32.367Q-14.681 32.784-14.659 33.126Q-14.637 33.468-14.548 33.816Q-14.459 34.165-14.254 34.396Q-14.049 34.627-13.714 34.627M-11.007 34.292Q-11.007 34.124-10.884 34.001Q-10.761 33.878-10.586 33.878Q-10.419 33.878-10.296 34.001Q-10.173 34.124-10.173 34.292Q-10.173 34.466-10.296 34.589Q-10.419 34.712-10.586 34.712Q-10.761 34.712-10.884 34.589Q-11.007 34.466-11.007 34.292M-7.469 34.852Q-8.105 34.852-8.469 34.507Q-8.833 34.162-8.968 33.637Q-9.103 33.112-9.103 32.487Q-9.103 31.462-8.747 30.763Q-8.392 30.064-7.469 30.064Q-6.543 30.064-6.191 30.763Q-5.839 31.462-5.839 32.487Q-5.839 33.112-5.974 33.637Q-6.109 34.162-6.471 34.507Q-6.833 34.852-7.469 34.852M-7.469 34.627Q-7.032 34.627-6.818 34.252Q-6.604 33.878-6.555 33.411Q-6.505 32.945-6.505 32.367Q-6.505 31.814-6.555 31.386Q-6.604 30.959-6.816 30.624Q-7.028 30.289-7.469 30.289Q-7.811 30.289-8.014 30.496Q-8.218 30.703-8.305 31.015Q-8.392 31.328-8.414 31.644Q-8.436 31.961-8.436 32.367Q-8.436 32.784-8.414 33.126Q-8.392 33.468-8.303 33.816Q-8.214 34.165-8.009 34.396Q-7.804 34.627-7.469 34.627M-2.151 34.712L-4.680 34.712L-4.680 34.432Q-3.713 34.432-3.713 34.223L-3.713 30.604Q-4.106 30.792-4.728 30.792L-4.728 30.511Q-4.311 30.511-3.947 30.410Q-3.583 30.310-3.327 30.064L-3.200 30.064Q-3.135 30.081-3.118 30.149L-3.118 34.223Q-3.118 34.432-2.151 34.432\" fill=\"var(--tk-warn)\" stroke=\"var(--tk-warn)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fsvg>\u003Cfigcaption class=\"tikz-cap\">The per-step multiplier compounds geometrically over \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6944em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathnormal\" style=\"margin-right:0.0315em;\">k\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> steps back. With a factor of \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6444em;\">\u003C\u002Fspan>\u003Cspan class=\"mord\">0.6\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> per hop the gradient scale is \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.8491em;\">\u003C\u002Fspan>\u003Cspan class=\"mord\">0.\u003C\u002Fspan>\u003Cspan class=\"mord\">\u003Cspan class=\"mord\">6\u003C\u002Fspan>\u003Cspan class=\"msupsub\">\u003Cspan class=\"vlist-t\">\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.8491em;\">\u003Cspan style=\"top:-3.063em;margin-right:0.05em;\">\u003Cspan class=\"pstrut\" style=\"height:2.7em;\">\u003C\u002Fspan>\u003Cspan class=\"sizing reset-size6 size3 mtight\">\u003Cspan class=\"mord mathnormal mtight\" style=\"margin-right:0.0315em;\">k\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>: it falls below \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6444em;\">\u003C\u002Fspan>\u003Cspan class=\"mord\">0.01\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> by about eight steps, so context more than a few words back gets a negligible learning signal.\u003C\u002Ffigcaption>","\u003Csvg style=\"width:100%;max-width:426.519px;height:auto\" xmlns=\"http:\u002F\u002Fwww.w3.org\u002F2000\u002Fsvg\" viewBox=\"-75 -75 319.890 128.627\">\u003Cg stroke=\"currentColor\" style=\"stroke-miterlimit:10;stroke-width:.4\">\u003Cpath fill=\"none\" d=\"M-40.322 37.557h237.003\"\u002F>\u003Cpath stroke=\"none\" d=\"m198.681 37.557-3.2-1.6 1.2 1.6-1.2 1.6\"\u002F>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(242.536 1.75)\">\u003Cpath d=\"M-40.008 37.550L-40.008 36.487Q-40.008 36.463-39.980 36.436Q-39.953 36.409-39.929 36.409L-39.820 36.409Q-39.755 36.409-39.741 36.467Q-39.645 36.901-39.399 37.152Q-39.153 37.403-38.739 37.403Q-38.398 37.403-38.145 37.270Q-37.892 37.137-37.892 36.829Q-37.892 36.672-37.986 36.557Q-38.080 36.443-38.218 36.374Q-38.357 36.306-38.524 36.268L-39.105 36.169Q-39.461 36.101-39.734 35.880Q-40.008 35.660-40.008 35.318Q-40.008 35.069-39.896 34.894Q-39.785 34.720-39.599 34.621Q-39.413 34.522-39.197 34.479Q-38.982 34.436-38.739 34.436Q-38.326 34.436-38.046 34.618L-37.830 34.443Q-37.820 34.440-37.813 34.438Q-37.806 34.436-37.796 34.436L-37.745 34.436Q-37.718 34.436-37.694 34.460Q-37.670 34.484-37.670 34.512L-37.670 35.359Q-37.670 35.380-37.694 35.407Q-37.718 35.434-37.745 35.434L-37.858 35.434Q-37.885 35.434-37.911 35.409Q-37.936 35.383-37.936 35.359Q-37.936 35.123-38.042 34.959Q-38.148 34.795-38.331 34.713Q-38.514 34.631-38.746 34.631Q-39.074 34.631-39.331 34.734Q-39.587 34.836-39.587 35.113Q-39.587 35.308-39.404 35.417Q-39.221 35.527-38.992 35.568L-38.418 35.674Q-38.172 35.722-37.958 35.850Q-37.745 35.978-37.608 36.181Q-37.471 36.385-37.471 36.634Q-37.471 37.147-37.837 37.386Q-38.203 37.625-38.739 37.625Q-39.235 37.625-39.567 37.331L-39.833 37.605Q-39.854 37.625-39.881 37.625L-39.929 37.625Q-39.953 37.625-39.980 37.598Q-40.008 37.571-40.008 37.550M-36.316 36.716L-36.316 34.819L-36.955 34.819L-36.955 34.597Q-36.637 34.597-36.420 34.387Q-36.203 34.177-36.103 33.867Q-36.002 33.558-36.002 33.250L-35.735 33.250L-35.735 34.539L-34.658 34.539L-34.658 34.819L-35.735 34.819L-35.735 36.703Q-35.735 36.979-35.631 37.178Q-35.527 37.376-35.267 37.376Q-35.110 37.376-35.004 37.272Q-34.898 37.167-34.848 37.014Q-34.799 36.860-34.799 36.703L-34.799 36.289L-34.532 36.289L-34.532 36.716Q-34.532 36.942-34.631 37.152Q-34.730 37.362-34.915 37.494Q-35.099 37.625-35.328 37.625Q-35.766 37.625-36.041 37.388Q-36.316 37.150-36.316 36.716M-33.763 36.022Q-33.763 35.701-33.638 35.412Q-33.513 35.123-33.288 34.900Q-33.062 34.676-32.767 34.556Q-32.471 34.436-32.153 34.436Q-31.825 34.436-31.563 34.536Q-31.302 34.635-31.126 34.817Q-30.950 35-30.856 35.258Q-30.762 35.516-30.762 35.848Q-30.762 35.940-30.844 35.961L-33.100 35.961L-33.100 36.022Q-33.100 36.610-32.816 36.993Q-32.532 37.376-31.965 37.376Q-31.644 37.376-31.375 37.183Q-31.107 36.990-31.018 36.675Q-31.011 36.634-30.936 36.620L-30.844 36.620Q-30.762 36.644-30.762 36.716Q-30.762 36.723-30.769 36.750Q-30.882 37.147-31.252 37.386Q-31.623 37.625-32.047 37.625Q-32.485 37.625-32.885 37.417Q-33.284 37.208-33.524 36.841Q-33.763 36.474-33.763 36.022M-33.093 35.752L-31.278 35.752Q-31.278 35.475-31.375 35.223Q-31.473 34.970-31.671 34.814Q-31.869 34.659-32.153 34.659Q-32.430 34.659-32.644 34.817Q-32.857 34.976-32.975 35.231Q-33.093 35.486-33.093 35.752M-28.530 38.914L-30.160 38.914L-30.160 38.634Q-29.931 38.634-29.783 38.599Q-29.634 38.565-29.634 38.425L-29.634 35.079Q-29.634 34.908-29.771 34.867Q-29.907 34.826-30.160 34.826L-30.160 34.546L-29.080 34.471L-29.080 34.877Q-28.858 34.676-28.571 34.573Q-28.284 34.471-27.976 34.471Q-27.549 34.471-27.185 34.684Q-26.821 34.898-26.607 35.262Q-26.394 35.626-26.394 36.046Q-26.394 36.491-26.633 36.855Q-26.872 37.219-27.265 37.422Q-27.658 37.625-28.103 37.625Q-28.369 37.625-28.617 37.525Q-28.865 37.424-29.053 37.243L-29.053 38.425Q-29.053 38.562-28.904 38.598Q-28.756 38.634-28.530 38.634L-28.530 38.914M-29.053 35.226L-29.053 36.836Q-28.920 37.089-28.677 37.246Q-28.434 37.403-28.157 37.403Q-27.829 37.403-27.576 37.202Q-27.323 37-27.190 36.682Q-27.057 36.364-27.057 36.046Q-27.057 35.817-27.122 35.588Q-27.187 35.359-27.315 35.161Q-27.443 34.963-27.638 34.843Q-27.833 34.724-28.065 34.724Q-28.359 34.724-28.627 34.853Q-28.896 34.983-29.053 35.226M-25.758 37.550L-25.758 36.487Q-25.758 36.463-25.731 36.436Q-25.703 36.409-25.679 36.409L-25.570 36.409Q-25.505 36.409-25.491 36.467Q-25.396 36.901-25.150 37.152Q-24.904 37.403-24.490 37.403Q-24.148 37.403-23.895 37.270Q-23.642 37.137-23.642 36.829Q-23.642 36.672-23.736 36.557Q-23.830 36.443-23.969 36.374Q-24.107 36.306-24.275 36.268L-24.856 36.169Q-25.211 36.101-25.485 35.880Q-25.758 35.660-25.758 35.318Q-25.758 35.069-25.647 34.894Q-25.536 34.720-25.350 34.621Q-25.163 34.522-24.948 34.479Q-24.733 34.436-24.490 34.436Q-24.076 34.436-23.796 34.618L-23.581 34.443Q-23.571 34.440-23.564 34.438Q-23.557 34.436-23.547 34.436L-23.495 34.436Q-23.468 34.436-23.444 34.460Q-23.420 34.484-23.420 34.512L-23.420 35.359Q-23.420 35.380-23.444 35.407Q-23.468 35.434-23.495 35.434L-23.608 35.434Q-23.635 35.434-23.661 35.409Q-23.687 35.383-23.687 35.359Q-23.687 35.123-23.793 34.959Q-23.899 34.795-24.082 34.713Q-24.264 34.631-24.497 34.631Q-24.825 34.631-25.081 34.734Q-25.338 34.836-25.338 35.113Q-25.338 35.308-25.155 35.417Q-24.972 35.527-24.743 35.568L-24.169 35.674Q-23.923 35.722-23.709 35.850Q-23.495 35.978-23.359 36.181Q-23.222 36.385-23.222 36.634Q-23.222 37.147-23.588 37.386Q-23.953 37.625-24.490 37.625Q-24.986 37.625-25.317 37.331L-25.584 37.605Q-25.604 37.625-25.632 37.625L-25.679 37.625Q-25.703 37.625-25.731 37.598Q-25.758 37.571-25.758 37.550\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(242.536 1.75)\">\u003Cpath d=\"M-19.073 37.557L-19.340 37.557L-19.340 33.449Q-19.340 33.179-19.447 33.117Q-19.555 33.056-19.866 33.056L-19.866 32.775L-18.786 32.700L-18.786 34.870Q-18.577 34.679-18.292 34.575Q-18.006 34.471-17.709 34.471Q-17.391 34.471-17.094 34.592Q-16.797 34.713-16.574 34.929Q-16.352 35.144-16.226 35.429Q-16.099 35.715-16.099 36.046Q-16.099 36.491-16.339 36.855Q-16.578 37.219-16.971 37.422Q-17.364 37.625-17.808 37.625Q-18.003 37.625-18.193 37.569Q-18.382 37.513-18.543 37.408Q-18.704 37.304-18.844 37.143L-19.073 37.557M-18.758 35.212L-18.758 36.829Q-18.622 37.089-18.381 37.246Q-18.140 37.403-17.863 37.403Q-17.569 37.403-17.357 37.296Q-17.145 37.188-17.012 36.996Q-16.879 36.805-16.820 36.566Q-16.762 36.327-16.762 36.046Q-16.762 35.687-16.856 35.383Q-16.950 35.079-17.178 34.886Q-17.405 34.693-17.771 34.693Q-18.071 34.693-18.338 34.829Q-18.605 34.966-18.758 35.212M-15.405 36.829Q-15.405 36.497-15.182 36.270Q-14.958 36.043-14.614 35.915Q-14.271 35.786-13.898 35.734Q-13.526 35.681-13.221 35.681L-13.221 35.428Q-13.221 35.223-13.329 35.043Q-13.437 34.864-13.618 34.761Q-13.799 34.659-14.007 34.659Q-14.414 34.659-14.650 34.751Q-14.561 34.788-14.515 34.872Q-14.469 34.956-14.469 35.058Q-14.469 35.154-14.515 35.233Q-14.561 35.311-14.642 35.356Q-14.722 35.400-14.811 35.400Q-14.961 35.400-15.062 35.303Q-15.163 35.205-15.163 35.058Q-15.163 34.436-14.007 34.436Q-13.796 34.436-13.546 34.500Q-13.297 34.563-13.095 34.682Q-12.893 34.802-12.767 34.987Q-12.640 35.171-12.640 35.414L-12.640 36.990Q-12.640 37.106-12.579 37.202Q-12.517 37.297-12.404 37.297Q-12.295 37.297-12.230 37.203Q-12.165 37.109-12.165 36.990L-12.165 36.542L-11.899 36.542L-11.899 36.990Q-11.899 37.260-12.126 37.425Q-12.353 37.591-12.633 37.591Q-12.842 37.591-12.979 37.437Q-13.115 37.284-13.139 37.068Q-13.286 37.335-13.568 37.480Q-13.850 37.625-14.175 37.625Q-14.452 37.625-14.736 37.550Q-15.019 37.475-15.212 37.296Q-15.405 37.116-15.405 36.829M-14.790 36.829Q-14.790 37.003-14.689 37.133Q-14.589 37.263-14.433 37.333Q-14.277 37.403-14.113 37.403Q-13.895 37.403-13.686 37.306Q-13.478 37.208-13.350 37.027Q-13.221 36.846-13.221 36.620L-13.221 35.892Q-13.546 35.892-13.912 35.983Q-14.277 36.074-14.534 36.286Q-14.790 36.497-14.790 36.829M-11.482 36.046Q-11.482 35.718-11.347 35.417Q-11.212 35.117-10.976 34.896Q-10.740 34.676-10.436 34.556Q-10.131 34.436-9.807 34.436Q-9.301 34.436-8.952 34.539Q-8.604 34.641-8.604 35.017Q-8.604 35.164-8.701 35.265Q-8.798 35.366-8.945 35.366Q-9.099 35.366-9.198 35.267Q-9.298 35.168-9.298 35.017Q-9.298 34.829-9.157 34.737Q-9.359 34.686-9.800 34.686Q-10.155 34.686-10.384 34.882Q-10.613 35.079-10.714 35.388Q-10.815 35.698-10.815 36.046Q-10.815 36.395-10.689 36.701Q-10.562 37.007-10.308 37.191Q-10.053 37.376-9.697 37.376Q-9.475 37.376-9.291 37.292Q-9.106 37.208-8.971 37.053Q-8.836 36.897-8.778 36.689Q-8.764 36.634-8.710 36.634L-8.597 36.634Q-8.566 36.634-8.544 36.658Q-8.522 36.682-8.522 36.716L-8.522 36.737Q-8.607 37.024-8.795 37.222Q-8.983 37.420-9.248 37.523Q-9.513 37.625-9.807 37.625Q-10.237 37.625-10.625 37.419Q-11.013 37.212-11.247 36.849Q-11.482 36.487-11.482 36.046\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(242.536 1.75)\">\u003Cpath d=\"M-6.540 37.557L-8.123 37.557L-8.123 37.277Q-7.894 37.277-7.745 37.243Q-7.597 37.208-7.597 37.068L-7.597 33.449Q-7.597 33.179-7.704 33.117Q-7.812 33.056-8.123 33.056L-8.123 32.775L-7.043 32.700L-7.043 35.988L-6.058 35.219Q-5.853 35.082-5.853 34.932Q-5.853 34.888-5.894 34.853Q-5.935 34.819-5.980 34.819L-5.980 34.539L-4.616 34.539L-4.616 34.819Q-5.105 34.819-5.624 35.219L-6.181 35.653L-5.204 36.877Q-5.002 37.123-4.869 37.200Q-4.736 37.277-4.449 37.277L-4.449 37.557L-5.881 37.557L-5.881 37.277Q-5.693 37.277-5.693 37.164Q-5.693 37.068-5.847 36.877L-6.581 35.968L-7.063 36.347L-7.063 37.068Q-7.063 37.205-6.915 37.241Q-6.766 37.277-6.540 37.277\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-40.322 37.557v-94.739\"\u002F>\u003Cpath stroke=\"none\" d=\"m-40.322-59.182-1.6 3.2 1.6-1.2 1.6 1.2\"\u002F>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(-21.949 -101.633)\">\u003Cpath d=\"M-40.049 38.090Q-40.049 37.844-39.852 37.660Q-39.655 37.475-39.399 37.396Q-39.536 37.284-39.608 37.123Q-39.679 36.962-39.679 36.781Q-39.679 36.460-39.468 36.214Q-39.802 35.916-39.802 35.506Q-39.802 35.045-39.413 34.758Q-39.023 34.471-38.545 34.471Q-38.073 34.471-37.738 34.717Q-37.564 34.563-37.353 34.481Q-37.143 34.399-36.914 34.399Q-36.750 34.399-36.629 34.506Q-36.508 34.614-36.508 34.778Q-36.508 34.874-36.579 34.946Q-36.651 35.017-36.743 35.017Q-36.843 35.017-36.913 34.944Q-36.983 34.870-36.983 34.771Q-36.983 34.717-36.969 34.686L-36.962 34.672Q-36.955 34.652-36.947 34.641Q-36.938 34.631-36.935 34.624Q-37.290 34.624-37.577 34.847Q-37.290 35.140-37.290 35.506Q-37.290 35.821-37.475 36.053Q-37.659 36.286-37.948 36.414Q-38.237 36.542-38.545 36.542Q-38.746 36.542-38.938 36.492Q-39.129 36.443-39.307 36.333Q-39.399 36.460-39.399 36.603Q-39.399 36.785-39.271 36.920Q-39.143 37.055-38.958 37.055L-38.326 37.055Q-37.878 37.055-37.509 37.126Q-37.140 37.198-36.880 37.427Q-36.620 37.656-36.620 38.090Q-36.620 38.411-36.916 38.613Q-37.212 38.815-37.615 38.904Q-38.018 38.993-38.333 38.993Q-38.651 38.993-39.054 38.904Q-39.457 38.815-39.753 38.613Q-40.049 38.411-40.049 38.090M-39.594 38.090Q-39.594 38.319-39.375 38.468Q-39.156 38.617-38.864 38.685Q-38.572 38.753-38.333 38.753Q-38.169 38.753-37.960 38.717Q-37.752 38.682-37.545 38.601Q-37.338 38.521-37.207 38.393Q-37.075 38.265-37.075 38.090Q-37.075 37.738-37.456 37.644Q-37.837 37.550-38.340 37.550L-38.958 37.550Q-39.197 37.550-39.396 37.701Q-39.594 37.851-39.594 38.090M-38.545 36.303Q-37.878 36.303-37.878 35.506Q-37.878 34.706-38.545 34.706Q-39.215 34.706-39.215 35.506Q-39.215 36.303-38.545 36.303M-34.276 37.557L-36.012 37.557L-36.012 37.277Q-35.783 37.277-35.634 37.243Q-35.486 37.208-35.486 37.068L-35.486 35.219Q-35.486 34.949-35.593 34.888Q-35.701 34.826-36.012 34.826L-36.012 34.546L-34.983 34.471L-34.983 35.178Q-34.853 34.870-34.611 34.671Q-34.368 34.471-34.050 34.471Q-33.831 34.471-33.660 34.595Q-33.489 34.720-33.489 34.932Q-33.489 35.069-33.589 35.168Q-33.688 35.267-33.821 35.267Q-33.958 35.267-34.057 35.168Q-34.156 35.069-34.156 34.932Q-34.156 34.792-34.057 34.693Q-34.347 34.693-34.547 34.889Q-34.747 35.086-34.840 35.380Q-34.932 35.674-34.932 35.954L-34.932 37.068Q-34.932 37.277-34.276 37.277L-34.276 37.557M-32.847 36.829Q-32.847 36.497-32.623 36.270Q-32.399 36.043-32.056 35.915Q-31.712 35.786-31.340 35.734Q-30.967 35.681-30.663 35.681L-30.663 35.428Q-30.663 35.223-30.770 35.043Q-30.878 34.864-31.059 34.761Q-31.240 34.659-31.449 34.659Q-31.856 34.659-32.092 34.751Q-32.003 34.788-31.957 34.872Q-31.910 34.956-31.910 35.058Q-31.910 35.154-31.957 35.233Q-32.003 35.311-32.083 35.356Q-32.163 35.400-32.252 35.400Q-32.403 35.400-32.503 35.303Q-32.604 35.205-32.604 35.058Q-32.604 34.436-31.449 34.436Q-31.237 34.436-30.988 34.500Q-30.738 34.563-30.536 34.682Q-30.335 34.802-30.208 34.987Q-30.082 35.171-30.082 35.414L-30.082 36.990Q-30.082 37.106-30.020 37.202Q-29.959 37.297-29.846 37.297Q-29.737 37.297-29.672 37.203Q-29.607 37.109-29.607 36.990L-29.607 36.542L-29.340 36.542L-29.340 36.990Q-29.340 37.260-29.567 37.425Q-29.795 37.591-30.075 37.591Q-30.283 37.591-30.420 37.437Q-30.557 37.284-30.581 37.068Q-30.728 37.335-31.010 37.480Q-31.292 37.625-31.616 37.625Q-31.893 37.625-32.177 37.550Q-32.461 37.475-32.654 37.296Q-32.847 37.116-32.847 36.829M-32.232 36.829Q-32.232 37.003-32.131 37.133Q-32.030 37.263-31.874 37.333Q-31.719 37.403-31.555 37.403Q-31.336 37.403-31.128 37.306Q-30.919 37.208-30.791 37.027Q-30.663 36.846-30.663 36.620L-30.663 35.892Q-30.988 35.892-31.353 35.983Q-31.719 36.074-31.975 36.286Q-32.232 36.497-32.232 36.829M-28.923 36.046Q-28.923 35.708-28.783 35.417Q-28.643 35.127-28.398 34.913Q-28.154 34.700-27.850 34.585Q-27.546 34.471-27.221 34.471Q-26.951 34.471-26.688 34.570Q-26.425 34.669-26.233 34.847L-26.233 33.449Q-26.233 33.179-26.341 33.117Q-26.448 33.056-26.760 33.056L-26.760 32.775L-25.683 32.700L-25.683 36.884Q-25.683 37.072-25.628 37.155Q-25.573 37.239-25.473 37.258Q-25.372 37.277-25.156 37.277L-25.156 37.557L-26.264 37.625L-26.264 37.208Q-26.681 37.625-27.306 37.625Q-27.737 37.625-28.110 37.413Q-28.482 37.202-28.703 36.841Q-28.923 36.480-28.923 36.046M-27.248 37.403Q-27.040 37.403-26.853 37.331Q-26.667 37.260-26.513 37.123Q-26.360 36.986-26.264 36.808L-26.264 35.199Q-26.349 35.052-26.495 34.932Q-26.640 34.812-26.809 34.753Q-26.978 34.693-27.159 34.693Q-27.720 34.693-27.988 35.082Q-28.257 35.472-28.257 36.053Q-28.257 36.624-28.022 37.014Q-27.788 37.403-27.248 37.403M-22.890 37.557L-24.442 37.557L-24.442 37.277Q-24.217 37.277-24.068 37.243Q-23.919 37.208-23.919 37.068L-23.919 35.219Q-23.919 35.031-23.967 34.947Q-24.015 34.864-24.112 34.845Q-24.210 34.826-24.422 34.826L-24.422 34.546L-23.365 34.471L-23.365 37.068Q-23.365 37.208-23.234 37.243Q-23.102 37.277-22.890 37.277L-22.890 37.557M-24.162 33.250Q-24.162 33.079-24.039 32.960Q-23.916 32.840-23.745 32.840Q-23.577 32.840-23.454 32.960Q-23.331 33.079-23.331 33.250Q-23.331 33.425-23.454 33.548Q-23.577 33.671-23.745 33.671Q-23.916 33.671-24.039 33.548Q-24.162 33.425-24.162 33.250M-22.285 36.022Q-22.285 35.701-22.161 35.412Q-22.036 35.123-21.810 34.900Q-21.585 34.676-21.289 34.556Q-20.993 34.436-20.676 34.436Q-20.347 34.436-20.086 34.536Q-19.824 34.635-19.648 34.817Q-19.472 35-19.378 35.258Q-19.284 35.516-19.284 35.848Q-19.284 35.940-19.366 35.961L-21.622 35.961L-21.622 36.022Q-21.622 36.610-21.339 36.993Q-21.055 37.376-20.488 37.376Q-20.166 37.376-19.898 37.183Q-19.630 36.990-19.541 36.675Q-19.534 36.634-19.459 36.620L-19.366 36.620Q-19.284 36.644-19.284 36.716Q-19.284 36.723-19.291 36.750Q-19.404 37.147-19.775 37.386Q-20.146 37.625-20.570 37.625Q-21.007 37.625-21.407 37.417Q-21.807 37.208-22.046 36.841Q-22.285 36.474-22.285 36.022M-21.615 35.752L-19.801 35.752Q-19.801 35.475-19.898 35.223Q-19.995 34.970-20.194 34.814Q-20.392 34.659-20.676 34.659Q-20.952 34.659-21.166 34.817Q-21.380 34.976-21.498 35.231Q-21.615 35.486-21.615 35.752M-17.015 37.557L-18.649 37.557L-18.649 37.277Q-18.420 37.277-18.271 37.243Q-18.122 37.208-18.122 37.068L-18.122 35.219Q-18.122 34.949-18.230 34.888Q-18.338 34.826-18.649 34.826L-18.649 34.546L-17.589 34.471L-17.589 35.120Q-17.418 34.812-17.114 34.641Q-16.810 34.471-16.465 34.471Q-15.959 34.471-15.675 34.694Q-15.391 34.918-15.391 35.414L-15.391 37.068Q-15.391 37.205-15.243 37.241Q-15.094 37.277-14.868 37.277L-14.868 37.557L-16.499 37.557L-16.499 37.277Q-16.270 37.277-16.121 37.243Q-15.972 37.208-15.972 37.068L-15.972 35.428Q-15.972 35.093-16.092 34.893Q-16.212 34.693-16.526 34.693Q-16.796 34.693-17.030 34.829Q-17.264 34.966-17.403 35.200Q-17.541 35.434-17.541 35.708L-17.541 37.068Q-17.541 37.205-17.391 37.241Q-17.240 37.277-17.015 37.277\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(-21.949 -101.633)\">\u003Cpath d=\"M-13.946 36.716L-13.946 34.819L-14.585 34.819L-14.585 34.597Q-14.267 34.597-14.050 34.387Q-13.833 34.177-13.733 33.867Q-13.632 33.558-13.632 33.250L-13.365 33.250L-13.365 34.539L-12.288 34.539L-12.288 34.819L-13.365 34.819L-13.365 36.703Q-13.365 36.979-13.261 37.178Q-13.157 37.376-12.897 37.376Q-12.740 37.376-12.634 37.272Q-12.528 37.167-12.478 37.014Q-12.429 36.860-12.429 36.703L-12.429 36.289L-12.162 36.289L-12.162 36.716Q-12.162 36.942-12.261 37.152Q-12.360 37.362-12.545 37.494Q-12.729 37.625-12.958 37.625Q-13.396 37.625-13.671 37.388Q-13.946 37.150-13.946 36.716\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(-21.949 -101.633)\">\u003Cpath d=\"M-8.653 37.550L-8.653 36.487Q-8.653 36.463-8.625 36.436Q-8.598 36.409-8.574 36.409L-8.465 36.409Q-8.400 36.409-8.386 36.467Q-8.290 36.901-8.044 37.152Q-7.798 37.403-7.384 37.403Q-7.043 37.403-6.790 37.270Q-6.537 37.137-6.537 36.829Q-6.537 36.672-6.631 36.557Q-6.725 36.443-6.863 36.374Q-7.002 36.306-7.169 36.268L-7.750 36.169Q-8.106 36.101-8.379 35.880Q-8.653 35.660-8.653 35.318Q-8.653 35.069-8.541 34.894Q-8.430 34.720-8.244 34.621Q-8.058 34.522-7.842 34.479Q-7.627 34.436-7.384 34.436Q-6.971 34.436-6.691 34.618L-6.475 34.443Q-6.465 34.440-6.458 34.438Q-6.451 34.436-6.441 34.436L-6.390 34.436Q-6.363 34.436-6.339 34.460Q-6.315 34.484-6.315 34.512L-6.315 35.359Q-6.315 35.380-6.339 35.407Q-6.363 35.434-6.390 35.434L-6.503 35.434Q-6.530 35.434-6.556 35.409Q-6.581 35.383-6.581 35.359Q-6.581 35.123-6.687 34.959Q-6.793 34.795-6.976 34.713Q-7.159 34.631-7.391 34.631Q-7.719 34.631-7.976 34.734Q-8.232 34.836-8.232 35.113Q-8.232 35.308-8.049 35.417Q-7.866 35.527-7.637 35.568L-7.063 35.674Q-6.817 35.722-6.603 35.850Q-6.390 35.978-6.253 36.181Q-6.116 36.385-6.116 36.634Q-6.116 37.147-6.482 37.386Q-6.848 37.625-7.384 37.625Q-7.880 37.625-8.212 37.331L-8.478 37.605Q-8.499 37.625-8.526 37.625L-8.574 37.625Q-8.598 37.625-8.625 37.598Q-8.653 37.571-8.653 37.550M-3.871 37.557L-5.423 37.557L-5.423 37.277Q-5.197 37.277-5.048 37.243Q-4.900 37.208-4.900 37.068L-4.900 35.219Q-4.900 35.031-4.947 34.947Q-4.995 34.864-5.093 34.845Q-5.190 34.826-5.402 34.826L-5.402 34.546L-4.346 34.471L-4.346 37.068Q-4.346 37.208-4.214 37.243Q-4.083 37.277-3.871 37.277L-3.871 37.557M-5.142 33.250Q-5.142 33.079-5.019 32.960Q-4.896 32.840-4.725 32.840Q-4.558 32.840-4.435 32.960Q-4.312 33.079-4.312 33.250Q-4.312 33.425-4.435 33.548Q-4.558 33.671-4.725 33.671Q-4.896 33.671-5.019 33.548Q-5.142 33.425-5.142 33.250M-0.494 37.557L-3.167 37.557Q-3.211 37.557-3.238 37.530Q-3.266 37.502-3.266 37.458L-3.266 37.390Q-3.266 37.349-3.238 37.318L-1.130 34.765L-1.769 34.765Q-2.193 34.765-2.425 34.831Q-2.657 34.898-2.777 35.108Q-2.897 35.318-2.897 35.739L-3.160 35.739L-3.078 34.539L-0.487 34.539Q-0.446 34.539-0.417 34.566Q-0.388 34.594-0.388 34.638L-0.388 34.686Q-0.388 34.730-0.412 34.758L-2.517 37.304L-1.837 37.304Q-1.509 37.304-1.292 37.268Q-1.075 37.232-0.907 37.082Q-0.767 36.945-0.714 36.721Q-0.661 36.497-0.634 36.169L-0.367 36.169L-0.494 37.557M0.282 36.022Q0.282 35.701 0.407 35.412Q0.532 35.123 0.757 34.900Q0.983 34.676 1.278 34.556Q1.574 34.436 1.892 34.436Q2.220 34.436 2.481 34.536Q2.743 34.635 2.919 34.817Q3.095 35 3.189 35.258Q3.283 35.516 3.283 35.848Q3.283 35.940 3.201 35.961L0.945 35.961L0.945 36.022Q0.945 36.610 1.229 36.993Q1.512 37.376 2.080 37.376Q2.401 37.376 2.669 37.183Q2.938 36.990 3.027 36.675Q3.033 36.634 3.109 36.620L3.201 36.620Q3.283 36.644 3.283 36.716Q3.283 36.723 3.276 36.750Q3.163 37.147 2.793 37.386Q2.422 37.625 1.998 37.625Q1.560 37.625 1.160 37.417Q0.761 37.208 0.521 36.841Q0.282 36.474 0.282 36.022M0.952 35.752L2.767 35.752Q2.767 35.475 2.669 35.223Q2.572 34.970 2.374 34.814Q2.176 34.659 1.892 34.659Q1.615 34.659 1.401 34.817Q1.188 34.976 1.070 35.231Q0.952 35.486 0.952 35.752\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"var(--tk-soft-accent)\" stroke=\"var(--tk-accent)\" d=\"M-19.836 37.557V-47.8h15.934v85.358ZM8.617 37.557v-55.482h15.934v55.482ZM37.07 37.557V1.422h15.933v36.135ZM65.523 37.557V14.226h15.933v23.331ZM93.975 37.557v-15.08h15.934v15.08ZM122.428 37.557V27.6h15.934v9.958ZM150.88 37.557v-6.26h15.934v6.26Zm15.934-6.26\"\u002F>\u003Cpath fill=\"none\" stroke=\"var(--tk-warn)\" d=\"M-11.869-47.8C16.584-22.195 45.037.568 73.49 14.225q42.68 13.657 85.359 17.072\" style=\"stroke-width:1.2\"\u002F>\u003Cg fill=\"var(--tk-warn)\" stroke=\"var(--tk-warn)\">\u003Cg fill=\"var(--tk-warn)\" stroke=\"none\" font-family=\"cmr8\" font-size=\"8\">\u003Cg transform=\"translate(206.97 -3.482)\">\u003Cpath d=\"M-38.283 37.526L-39.506 34.670Q-39.588 34.495-39.732 34.450Q-39.877 34.405-40.146 34.405L-40.146 34.108L-38.435 34.108L-38.435 34.405Q-38.857 34.405-38.857 34.588Q-38.857 34.623-38.842 34.670L-37.896 36.862L-37.056 34.885Q-37.017 34.807-37.017 34.717Q-37.017 34.577-37.123 34.491Q-37.228 34.405-37.369 34.405L-37.369 34.108L-36.017 34.108L-36.017 34.405Q-36.541 34.405-36.756 34.885L-37.881 37.526Q-37.943 37.635-38.049 37.635L-38.115 37.635Q-38.228 37.635-38.283 37.526\" fill=\"var(--tk-warn)\" stroke=\"var(--tk-warn)\" class=\"tikz-text\" style=\"stroke-width:0.240\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(206.97 -3.482)\">\u003Cpath d=\"M-35.972 36.725Q-35.972 36.241-35.570 35.946Q-35.167 35.651-34.617 35.532Q-34.066 35.412-33.574 35.412L-33.574 35.123Q-33.574 34.897-33.689 34.690Q-33.804 34.483-34.001 34.364Q-34.199 34.245-34.429 34.245Q-34.855 34.245-35.140 34.350Q-35.070 34.377-35.023 34.432Q-34.976 34.487-34.951 34.557Q-34.925 34.627-34.925 34.702Q-34.925 34.807-34.976 34.899Q-35.027 34.991-35.119 35.041Q-35.210 35.092-35.316 35.092Q-35.421 35.092-35.513 35.041Q-35.605 34.991-35.656 34.899Q-35.706 34.807-35.706 34.702Q-35.706 34.284-35.318 34.137Q-34.929 33.991-34.429 33.991Q-34.097 33.991-33.744 34.121Q-33.390 34.252-33.162 34.506Q-32.933 34.760-32.933 35.108L-32.933 36.909Q-32.933 37.041-32.861 37.151Q-32.788 37.260-32.660 37.260Q-32.535 37.260-32.466 37.155Q-32.398 37.049-32.398 36.909L-32.398 36.397L-32.117 36.397L-32.117 36.909Q-32.117 37.112-32.234 37.270Q-32.351 37.428-32.533 37.512Q-32.714 37.596-32.917 37.596Q-33.148 37.596-33.300 37.424Q-33.453 37.252-33.484 37.022Q-33.644 37.303-33.953 37.469Q-34.261 37.635-34.613 37.635Q-35.124 37.635-35.548 37.412Q-35.972 37.190-35.972 36.725M-35.285 36.725Q-35.285 37.010-35.058 37.196Q-34.831 37.381-34.538 37.381Q-34.292 37.381-34.068 37.264Q-33.843 37.147-33.708 36.944Q-33.574 36.741-33.574 36.487L-33.574 35.655Q-33.839 35.655-34.124 35.709Q-34.410 35.764-34.681 35.893Q-34.953 36.022-35.119 36.229Q-35.285 36.436-35.285 36.725M-29.894 37.557L-31.749 37.557L-31.749 37.260Q-31.476 37.260-31.308 37.213Q-31.140 37.166-31.140 36.998L-31.140 34.862Q-31.140 34.647-31.203 34.551Q-31.265 34.455-31.384 34.434Q-31.503 34.412-31.749 34.412L-31.749 34.116L-30.558 34.030L-30.558 34.764Q-30.445 34.549-30.251 34.381Q-30.058 34.213-29.820 34.121Q-29.581 34.030-29.328 34.030Q-28.160 34.030-28.160 35.108L-28.160 36.998Q-28.160 37.166-27.990 37.213Q-27.820 37.260-27.550 37.260L-27.550 37.557L-29.406 37.557L-29.406 37.260Q-29.132 37.260-28.964 37.213Q-28.796 37.166-28.796 36.998L-28.796 35.123Q-28.796 34.741-28.917 34.512Q-29.038 34.284-29.390 34.284Q-29.703 34.284-29.956 34.446Q-30.210 34.608-30.357 34.877Q-30.503 35.147-30.503 35.444L-30.503 36.998Q-30.503 37.166-30.333 37.213Q-30.163 37.260-29.894 37.260L-29.894 37.557M-25.245 37.557L-27.023 37.557L-27.023 37.260Q-26.749 37.260-26.581 37.213Q-26.413 37.166-26.413 36.998L-26.413 34.862Q-26.413 34.647-26.470 34.551Q-26.527 34.455-26.640 34.434Q-26.753 34.412-26.999 34.412L-26.999 34.116L-25.800 34.030L-25.800 36.998Q-25.800 37.166-25.654 37.213Q-25.507 37.260-25.245 37.260L-25.245 37.557M-26.687 32.635Q-26.687 32.444-26.552 32.313Q-26.417 32.182-26.222 32.182Q-26.101 32.182-25.997 32.245Q-25.894 32.307-25.831 32.411Q-25.769 32.514-25.769 32.635Q-25.769 32.830-25.900 32.965Q-26.031 33.100-26.222 33.100Q-26.421 33.100-26.554 32.967Q-26.687 32.834-26.687 32.635M-24.703 37.549L-24.703 36.327Q-24.703 36.299-24.671 36.268Q-24.640 36.237-24.617 36.237L-24.511 36.237Q-24.441 36.237-24.425 36.299Q-24.363 36.620-24.224 36.860Q-24.085 37.100-23.853 37.241Q-23.620 37.381-23.312 37.381Q-23.074 37.381-22.865 37.321Q-22.656 37.260-22.519 37.112Q-22.382 36.963-22.382 36.717Q-22.382 36.463-22.593 36.297Q-22.804 36.131-23.074 36.077L-23.695 35.963Q-24.101 35.885-24.402 35.629Q-24.703 35.373-24.703 34.998Q-24.703 34.631-24.501 34.409Q-24.300 34.186-23.976 34.088Q-23.652 33.991-23.312 33.991Q-22.847 33.991-22.550 34.198L-22.328 34.014Q-22.304 33.991-22.273 33.991L-22.222 33.991Q-22.191 33.991-22.163 34.018Q-22.136 34.045-22.136 34.077L-22.136 35.061Q-22.136 35.092-22.162 35.121Q-22.187 35.151-22.222 35.151L-22.328 35.151Q-22.363 35.151-22.390 35.123Q-22.417 35.096-22.417 35.061Q-22.417 34.662-22.669 34.442Q-22.921 34.221-23.320 34.221Q-23.675 34.221-23.958 34.344Q-24.242 34.467-24.242 34.772Q-24.242 34.991-24.040 35.123Q-23.839 35.256-23.593 35.299L-22.968 35.412Q-22.538 35.502-22.230 35.799Q-21.921 36.096-21.921 36.510Q-21.921 37.080-22.320 37.358Q-22.718 37.635-23.312 37.635Q-23.863 37.635-24.214 37.299L-24.511 37.612Q-24.535 37.635-24.570 37.635L-24.617 37.635Q-24.640 37.635-24.671 37.604Q-24.703 37.573-24.703 37.549M-19.464 37.557L-21.320 37.557L-21.320 37.260Q-21.046 37.260-20.878 37.213Q-20.710 37.166-20.710 36.998L-20.710 32.838Q-20.710 32.623-20.773 32.528Q-20.835 32.432-20.954 32.411Q-21.074 32.389-21.320 32.389L-21.320 32.092L-20.097 32.006L-20.097 34.709Q-19.972 34.498-19.785 34.348Q-19.597 34.198-19.370 34.114Q-19.144 34.030-18.898 34.030Q-17.730 34.030-17.730 35.108L-17.730 36.998Q-17.730 37.166-17.560 37.213Q-17.390 37.260-17.120 37.260L-17.120 37.557L-18.976 37.557L-18.976 37.260Q-18.703 37.260-18.535 37.213Q-18.367 37.166-18.367 36.998L-18.367 35.123Q-18.367 34.741-18.488 34.512Q-18.609 34.284-18.960 34.284Q-19.273 34.284-19.527 34.446Q-19.781 34.608-19.927 34.877Q-20.074 35.147-20.074 35.444L-20.074 36.998Q-20.074 37.166-19.904 37.213Q-19.734 37.260-19.464 37.260L-19.464 37.557M-16.675 35.803Q-16.675 35.323-16.443 34.907Q-16.210 34.491-15.800 34.241Q-15.390 33.991-14.913 33.991Q-14.183 33.991-13.785 34.432Q-13.386 34.873-13.386 35.604Q-13.386 35.709-13.480 35.733L-15.929 35.733L-15.929 35.803Q-15.929 36.213-15.808 36.569Q-15.687 36.924-15.415 37.141Q-15.144 37.358-14.714 37.358Q-14.351 37.358-14.054 37.129Q-13.757 36.901-13.656 36.549Q-13.648 36.502-13.562 36.487L-13.480 36.487Q-13.386 36.514-13.386 36.596Q-13.386 36.604-13.394 36.635Q-13.456 36.862-13.595 37.045Q-13.734 37.229-13.925 37.362Q-14.117 37.495-14.335 37.565Q-14.554 37.635-14.792 37.635Q-15.163 37.635-15.501 37.498Q-15.839 37.362-16.107 37.110Q-16.374 36.858-16.525 36.518Q-16.675 36.178-16.675 35.803M-15.921 35.495L-13.960 35.495Q-13.960 35.190-14.062 34.899Q-14.163 34.608-14.380 34.426Q-14.597 34.245-14.913 34.245Q-15.214 34.245-15.445 34.432Q-15.675 34.620-15.798 34.911Q-15.921 35.202-15.921 35.495M-12.855 37.549L-12.855 36.327Q-12.855 36.299-12.824 36.268Q-12.792 36.237-12.769 36.237L-12.663 36.237Q-12.593 36.237-12.578 36.299Q-12.515 36.620-12.376 36.860Q-12.238 37.100-12.005 37.241Q-11.773 37.381-11.464 37.381Q-11.226 37.381-11.017 37.321Q-10.808 37.260-10.671 37.112Q-10.535 36.963-10.535 36.717Q-10.535 36.463-10.745 36.297Q-10.956 36.131-11.226 36.077L-11.847 35.963Q-12.253 35.885-12.554 35.629Q-12.855 35.373-12.855 34.998Q-12.855 34.631-12.654 34.409Q-12.453 34.186-12.128 34.088Q-11.804 33.991-11.464 33.991Q-10.999 33.991-10.703 34.198L-10.480 34.014Q-10.456 33.991-10.425 33.991L-10.374 33.991Q-10.343 33.991-10.316 34.018Q-10.288 34.045-10.288 34.077L-10.288 35.061Q-10.288 35.092-10.314 35.121Q-10.339 35.151-10.374 35.151L-10.480 35.151Q-10.515 35.151-10.542 35.123Q-10.570 35.096-10.570 35.061Q-10.570 34.662-10.822 34.442Q-11.074 34.221-11.472 34.221Q-11.828 34.221-12.111 34.344Q-12.394 34.467-12.394 34.772Q-12.394 34.991-12.193 35.123Q-11.992 35.256-11.745 35.299L-11.120 35.412Q-10.691 35.502-10.382 35.799Q-10.074 36.096-10.074 36.510Q-10.074 37.080-10.472 37.358Q-10.870 37.635-11.464 37.635Q-12.015 37.635-12.367 37.299L-12.663 37.612Q-12.687 37.635-12.722 37.635L-12.769 37.635Q-12.792 37.635-12.824 37.604Q-12.855 37.573-12.855 37.549\" fill=\"var(--tk-warn)\" stroke=\"var(--tk-warn)\" class=\"tikz-text\" style=\"stroke-width:0.240\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003Cg transform=\"translate(26.46 9.467)\">\u003Cpath d=\"M-36.996 37.557L-39.526 37.557L-39.526 37.277Q-38.558 37.277-38.558 37.068L-38.558 33.449Q-38.951 33.637-39.573 33.637L-39.573 33.356Q-39.156 33.356-38.792 33.255Q-38.428 33.155-38.172 32.909L-38.046 32.909Q-37.981 32.926-37.964 32.994L-37.964 37.068Q-37.964 37.277-36.996 37.277\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(111.818 9.467)\">\u003Cpath d=\"M-38.005 36.409L-40.049 36.409L-40.049 36.128L-37.718 32.956Q-37.683 32.909-37.618 32.909L-37.482 32.909Q-37.437 32.909-37.410 32.936Q-37.383 32.963-37.383 33.008L-37.383 36.128L-36.620 36.128L-36.620 36.409L-37.383 36.409L-37.383 37.068Q-37.383 37.277-36.627 37.277L-36.627 37.557L-38.760 37.557L-38.760 37.277Q-38.005 37.277-38.005 37.068L-38.005 36.409M-37.957 33.684L-39.748 36.128L-37.957 36.128\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(197.176 9.467)\">\u003Cpath d=\"M-38.886 37.349Q-38.886 36.843-38.757 36.335Q-38.627 35.828-38.389 35.366Q-38.152 34.905-37.817 34.484L-37.171 33.671L-37.984 33.671Q-38.569 33.671-38.965 33.679Q-39.362 33.688-39.385 33.708Q-39.488 33.825-39.567 34.351L-39.833 34.351L-39.587 32.827L-39.321 32.827L-39.321 32.847Q-39.321 32.915-39.245 32.958Q-39.170 33.001-39.092 33.008Q-38.900 33.032-38.705 33.038Q-38.510 33.045-38.319 33.047Q-38.128 33.049-37.929 33.049L-36.508 33.049L-36.508 33.237Q-36.518 33.285-36.528 33.295L-37.584 34.618Q-37.803 34.891-37.926 35.204Q-38.049 35.516-38.107 35.865Q-38.165 36.214-38.179 36.545Q-38.193 36.877-38.193 37.349Q-38.193 37.499-38.292 37.598Q-38.391 37.697-38.538 37.697Q-38.688 37.697-38.787 37.598Q-38.886 37.499-38.886 37.349\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fsvg>\u003Cfigcaption class=\"tikz-cap\">The gradient's magnitude as it is propagated back through time, one recurrent hop per step. Multiplying by a factor below one at each hop shrinks it geometrically, so distant steps get almost no learning signal — the vanishing-gradient problem.\u003C\u002Ffigcaption>","\u003Csvg style=\"width:100%;max-width:513.746px;height:auto\" xmlns=\"http:\u002F\u002Fwww.w3.org\u002F2000\u002Fsvg\" viewBox=\"-75 -75 385.309 157.200\">\u003Cg stroke=\"currentColor\" style=\"stroke-miterlimit:10;stroke-width:.4\">\u003Cpath fill=\"none\" d=\"M-58.778-37.926h31.298v-17.072h-31.298Z\"\u002F>\u003Cg transform=\"translate(-9.806 -117.751)\">\u003Cpath d=\"M-42.815 71.528Q-42.815 71.200-42.680 70.899Q-42.545 70.599-42.309 70.378Q-42.073 70.158-41.769 70.038Q-41.464 69.918-41.140 69.918Q-40.634 69.918-40.285 70.021Q-39.937 70.123-39.937 70.499Q-39.937 70.646-40.034 70.747Q-40.131 70.848-40.278 70.848Q-40.432 70.848-40.531 70.749Q-40.630 70.650-40.630 70.499Q-40.630 70.311-40.490 70.219Q-40.692 70.168-41.133 70.168Q-41.488 70.168-41.717 70.364Q-41.946 70.561-42.047 70.870Q-42.148 71.180-42.148 71.528Q-42.148 71.877-42.022 72.183Q-41.895 72.489-41.640 72.673Q-41.386 72.858-41.030 72.858Q-40.808 72.858-40.624 72.774Q-40.439 72.690-40.304 72.535Q-40.169 72.379-40.111 72.171Q-40.097 72.116-40.043 72.116L-39.930 72.116Q-39.899 72.116-39.877 72.140Q-39.855 72.164-39.855 72.198L-39.855 72.219Q-39.940 72.506-40.128 72.704Q-40.316 72.902-40.581 73.005Q-40.846 73.107-41.140 73.107Q-41.570 73.107-41.958 72.901Q-42.346 72.694-42.580 72.331Q-42.815 71.969-42.815 71.528M-37.137 74.789Q-37.688 74.389-38.058 73.834Q-38.429 73.278-38.610 72.632Q-38.792 71.986-38.792 71.289Q-38.792 70.776-38.691 70.281Q-38.590 69.785-38.385 69.334Q-38.180 68.883-37.867 68.491Q-37.554 68.100-37.137 67.796Q-37.127 67.792-37.120 67.791Q-37.113 67.789-37.103 67.789L-37.035 67.789Q-37.001 67.789-36.978 67.813Q-36.956 67.837-36.956 67.874Q-36.956 67.919-36.983 67.936Q-37.332 68.237-37.585 68.621Q-37.838 69.006-37.990 69.447Q-38.142 69.888-38.214 70.344Q-38.286 70.800-38.286 71.289Q-38.286 72.290-37.976 73.177Q-37.667 74.064-36.983 74.649Q-36.956 74.666-36.956 74.710Q-36.956 74.748-36.978 74.772Q-37.001 74.796-37.035 74.796L-37.103 74.796Q-37.110 74.792-37.119 74.791Q-37.127 74.789-37.137 74.789M-35.620 72.198L-35.620 70.301L-36.259 70.301L-36.259 70.079Q-35.941 70.079-35.724 69.869Q-35.507 69.659-35.406 69.349Q-35.305 69.040-35.305 68.732L-35.039 68.732L-35.039 70.021L-33.962 70.021L-33.962 70.301L-35.039 70.301L-35.039 72.185Q-35.039 72.461-34.934 72.660Q-34.830 72.858-34.570 72.858Q-34.413 72.858-34.307 72.754Q-34.201 72.649-34.152 72.496Q-34.102 72.342-34.102 72.185L-34.102 71.771L-33.836 71.771L-33.836 72.198Q-33.836 72.424-33.935 72.634Q-34.034 72.844-34.218 72.976Q-34.403 73.107-34.632 73.107Q-35.069 73.107-35.345 72.870Q-35.620 72.632-35.620 72.198M-31.142 71.785L-33.200 71.785L-33.200 71.282L-31.142 71.282L-31.142 71.785M-27.328 73.039L-29.857 73.039L-29.857 72.759Q-28.890 72.759-28.890 72.550L-28.890 68.931Q-29.283 69.119-29.905 69.119L-29.905 68.838Q-29.488 68.838-29.124 68.737Q-28.760 68.637-28.504 68.391L-28.377 68.391Q-28.312 68.408-28.295 68.476L-28.295 72.550Q-28.295 72.759-27.328 72.759L-27.328 73.039M-26.036 74.796L-26.104 74.796Q-26.138 74.796-26.160 74.770Q-26.183 74.745-26.183 74.710Q-26.183 74.666-26.152 74.649Q-25.796 74.345-25.547 73.955Q-25.297 73.565-25.145 73.133Q-24.993 72.701-24.923 72.232Q-24.853 71.764-24.853 71.289Q-24.853 70.810-24.923 70.344Q-24.993 69.877-25.147 69.442Q-25.301 69.006-25.552 68.618Q-25.803 68.230-26.152 67.936Q-26.183 67.919-26.183 67.874Q-26.183 67.840-26.160 67.815Q-26.138 67.789-26.104 67.789L-26.036 67.789Q-26.025 67.789-26.017 67.791Q-26.008 67.792-25.998 67.796Q-25.455 68.196-25.082 68.749Q-24.710 69.303-24.528 69.949Q-24.347 70.595-24.347 71.289Q-24.347 71.990-24.528 72.637Q-24.710 73.285-25.084 73.839Q-25.458 74.393-25.998 74.789Q-26.008 74.789-26.017 74.791Q-26.025 74.792-26.036 74.796\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-58.778 21.824h31.298V4.752h-31.298Z\"\u002F>\u003Cg transform=\"translate(-10.236 -58)\">\u003Cpath d=\"M-41.133 73.039L-42.767 73.039L-42.767 72.759Q-42.538 72.759-42.389 72.725Q-42.240 72.690-42.240 72.550L-42.240 68.931Q-42.240 68.661-42.348 68.599Q-42.456 68.538-42.767 68.538L-42.767 68.257L-41.687 68.182L-41.687 70.568Q-41.581 70.383-41.403 70.241Q-41.225 70.100-41.017 70.026Q-40.808 69.953-40.583 69.953Q-40.077 69.953-39.793 70.176Q-39.509 70.400-39.509 70.896L-39.509 72.550Q-39.509 72.687-39.361 72.723Q-39.212 72.759-38.986 72.759L-38.986 73.039L-40.617 73.039L-40.617 72.759Q-40.388 72.759-40.239 72.725Q-40.090 72.690-40.090 72.550L-40.090 70.910Q-40.090 70.575-40.210 70.375Q-40.330 70.175-40.644 70.175Q-40.914 70.175-41.148 70.311Q-41.382 70.448-41.521 70.682Q-41.659 70.916-41.659 71.190L-41.659 72.550Q-41.659 72.687-41.509 72.723Q-41.358 72.759-41.133 72.759L-41.133 73.039M-36.269 74.789Q-36.819 74.389-37.190 73.834Q-37.561 73.278-37.742 72.632Q-37.923 71.986-37.923 71.289Q-37.923 70.776-37.823 70.281Q-37.722 69.785-37.517 69.334Q-37.312 68.883-36.999 68.491Q-36.686 68.100-36.269 67.796Q-36.259 67.792-36.252 67.791Q-36.245 67.789-36.235 67.789L-36.167 67.789Q-36.132 67.789-36.110 67.813Q-36.088 67.837-36.088 67.874Q-36.088 67.919-36.115 67.936Q-36.464 68.237-36.717 68.621Q-36.970 69.006-37.122 69.447Q-37.274 69.888-37.346 70.344Q-37.418 70.800-37.418 71.289Q-37.418 72.290-37.108 73.177Q-36.799 74.064-36.115 74.649Q-36.088 74.666-36.088 74.710Q-36.088 74.748-36.110 74.772Q-36.132 74.796-36.167 74.796L-36.235 74.796Q-36.242 74.792-36.250 74.791Q-36.259 74.789-36.269 74.789M-34.752 72.198L-34.752 70.301L-35.391 70.301L-35.391 70.079Q-35.073 70.079-34.856 69.869Q-34.639 69.659-34.538 69.349Q-34.437 69.040-34.437 68.732L-34.171 68.732L-34.171 70.021L-33.094 70.021L-33.094 70.301L-34.171 70.301L-34.171 72.185Q-34.171 72.461-34.066 72.660Q-33.962 72.858-33.702 72.858Q-33.545 72.858-33.439 72.754Q-33.333 72.649-33.284 72.496Q-33.234 72.342-33.234 72.185L-33.234 71.771L-32.967 71.771L-32.967 72.198Q-32.967 72.424-33.066 72.634Q-33.166 72.844-33.350 72.976Q-33.535 73.107-33.764 73.107Q-34.201 73.107-34.476 72.870Q-34.752 72.632-34.752 72.198M-30.274 71.785L-32.332 71.785L-32.332 71.282L-30.274 71.282L-30.274 71.785M-26.460 73.039L-28.989 73.039L-28.989 72.759Q-28.022 72.759-28.022 72.550L-28.022 68.931Q-28.415 69.119-29.037 69.119L-29.037 68.838Q-28.620 68.838-28.256 68.737Q-27.892 68.637-27.635 68.391L-27.509 68.391Q-27.444 68.408-27.427 68.476L-27.427 72.550Q-27.427 72.759-26.460 72.759L-26.460 73.039M-25.168 74.796L-25.236 74.796Q-25.270 74.796-25.292 74.770Q-25.315 74.745-25.315 74.710Q-25.315 74.666-25.284 74.649Q-24.928 74.345-24.679 73.955Q-24.429 73.565-24.277 73.133Q-24.125 72.701-24.055 72.232Q-23.985 71.764-23.985 71.289Q-23.985 70.810-24.055 70.344Q-24.125 69.877-24.279 69.442Q-24.433 69.006-24.684 68.618Q-24.935 68.230-25.284 67.936Q-25.315 67.919-25.315 67.874Q-25.315 67.840-25.292 67.815Q-25.270 67.789-25.236 67.789L-25.168 67.789Q-25.157 67.789-25.149 67.791Q-25.140 67.792-25.130 67.796Q-24.587 68.196-24.214 68.749Q-23.841 69.303-23.660 69.949Q-23.479 70.595-23.479 71.289Q-23.479 71.990-23.660 72.637Q-23.841 73.285-24.216 73.839Q-24.590 74.393-25.130 74.789Q-25.140 74.789-25.149 74.791Q-25.157 74.792-25.168 74.796\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-58.778 70.194h31.298V53.122h-31.298Z\"\u002F>\u003Cg transform=\"translate(-6.788 -9.63)\">\u003Cpath d=\"M-41.632 73.039L-42.955 73.039L-42.955 72.759Q-42.394 72.759-42.015 72.359L-41.300 71.562L-42.213 70.513Q-42.350 70.366-42.498 70.334Q-42.647 70.301-42.914 70.301L-42.914 70.021L-41.413 70.021L-41.413 70.301Q-41.605 70.301-41.605 70.435Q-41.605 70.465-41.574 70.513L-40.979 71.197L-40.538 70.701Q-40.425 70.571-40.425 70.455Q-40.425 70.393-40.463 70.347Q-40.501 70.301-40.559 70.301L-40.559 70.021L-39.243 70.021L-39.243 70.301Q-39.803 70.301-40.183 70.701L-40.805 71.402L-39.810 72.550Q-39.711 72.649-39.610 72.694Q-39.509 72.738-39.398 72.748Q-39.287 72.759-39.109 72.759L-39.109 73.039L-40.603 73.039L-40.603 72.759Q-40.538 72.759-40.478 72.725Q-40.419 72.690-40.419 72.625Q-40.419 72.578-40.449 72.550L-41.126 71.764L-41.659 72.359Q-41.772 72.489-41.772 72.605Q-41.772 72.670-41.731 72.714Q-41.690 72.759-41.632 72.759L-41.632 73.039M-36.484 74.789Q-37.035 74.389-37.406 73.834Q-37.776 73.278-37.958 72.632Q-38.139 71.986-38.139 71.289Q-38.139 70.776-38.038 70.281Q-37.937 69.785-37.732 69.334Q-37.527 68.883-37.214 68.491Q-36.901 68.100-36.484 67.796Q-36.474 67.792-36.467 67.791Q-36.461 67.789-36.450 67.789L-36.382 67.789Q-36.348 67.789-36.326 67.813Q-36.303 67.837-36.303 67.874Q-36.303 67.919-36.331 67.936Q-36.679 68.237-36.932 68.621Q-37.185 69.006-37.337 69.447Q-37.489 69.888-37.561 70.344Q-37.633 70.800-37.633 71.289Q-37.633 72.290-37.324 73.177Q-37.014 74.064-36.331 74.649Q-36.303 74.666-36.303 74.710Q-36.303 74.748-36.326 74.772Q-36.348 74.796-36.382 74.796L-36.450 74.796Q-36.457 74.792-36.466 74.791Q-36.474 74.789-36.484 74.789M-34.967 72.198L-34.967 70.301L-35.606 70.301L-35.606 70.079Q-35.288 70.079-35.071 69.869Q-34.854 69.659-34.753 69.349Q-34.652 69.040-34.652 68.732L-34.386 68.732L-34.386 70.021L-33.309 70.021L-33.309 70.301L-34.386 70.301L-34.386 72.185Q-34.386 72.461-34.282 72.660Q-34.177 72.858-33.918 72.858Q-33.760 72.858-33.654 72.754Q-33.548 72.649-33.499 72.496Q-33.449 72.342-33.449 72.185L-33.449 71.771L-33.183 71.771L-33.183 72.198Q-33.183 72.424-33.282 72.634Q-33.381 72.844-33.566 72.976Q-33.750 73.107-33.979 73.107Q-34.417 73.107-34.692 72.870Q-34.967 72.632-34.967 72.198M-32.051 74.796L-32.120 74.796Q-32.154 74.796-32.176 74.770Q-32.198 74.745-32.198 74.710Q-32.198 74.666-32.168 74.649Q-31.812 74.345-31.563 73.955Q-31.313 73.565-31.161 73.133Q-31.009 72.701-30.939 72.232Q-30.869 71.764-30.869 71.289Q-30.869 70.810-30.939 70.344Q-31.009 69.877-31.163 69.442Q-31.316 69.006-31.568 68.618Q-31.819 68.230-32.168 67.936Q-32.198 67.919-32.198 67.874Q-32.198 67.840-32.176 67.815Q-32.154 67.789-32.120 67.789L-32.051 67.789Q-32.041 67.789-32.033 67.791Q-32.024 67.792-32.014 67.796Q-31.470 68.196-31.098 68.749Q-30.725 69.303-30.544 69.949Q-30.363 70.595-30.363 71.289Q-30.363 71.990-30.544 72.637Q-30.725 73.285-31.099 73.839Q-31.474 74.393-32.014 74.789Q-32.024 74.789-32.033 74.791Q-32.041 74.792-32.051 74.796\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M56.455-46.462a8.536 8.536 0 1 0-17.071 0 8.536 8.536 0 0 0 17.071 0Zm-8.536 0\"\u002F>\u003Cg transform=\"translate(88.948 -117.994)\">\u003Cpath d=\"M-41.632 73.039L-42.955 73.039L-42.955 72.759Q-42.394 72.759-42.015 72.359L-41.300 71.562L-42.213 70.513Q-42.350 70.366-42.498 70.334Q-42.647 70.301-42.914 70.301L-42.914 70.021L-41.413 70.021L-41.413 70.301Q-41.605 70.301-41.605 70.435Q-41.605 70.465-41.574 70.513L-40.979 71.197L-40.538 70.701Q-40.425 70.571-40.425 70.455Q-40.425 70.393-40.463 70.347Q-40.501 70.301-40.559 70.301L-40.559 70.021L-39.243 70.021L-39.243 70.301Q-39.803 70.301-40.183 70.701L-40.805 71.402L-39.810 72.550Q-39.711 72.649-39.610 72.694Q-39.509 72.738-39.398 72.748Q-39.287 72.759-39.109 72.759L-39.109 73.039L-40.603 73.039L-40.603 72.759Q-40.538 72.759-40.478 72.725Q-40.419 72.690-40.419 72.625Q-40.419 72.578-40.449 72.550L-41.126 71.764L-41.659 72.359Q-41.772 72.489-41.772 72.605Q-41.772 72.670-41.731 72.714Q-41.690 72.759-41.632 72.759\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M147.504-46.462a8.536 8.536 0 1 0-17.072 0 8.536 8.536 0 0 0 17.072 0Zm-8.536 0\"\u002F>\u003Cg transform=\"translate(179.028 -117.751)\">\u003Cpath d=\"M-40.231 73.719L-40.231 71.463L-42.480 71.463Q-42.548 71.453-42.594 71.407Q-42.640 71.361-42.640 71.289Q-42.640 71.145-42.480 71.122L-40.231 71.122L-40.231 68.866Q-40.220 68.797-40.174 68.751Q-40.128 68.705-40.056 68.705Q-39.913 68.705-39.889 68.866L-39.889 71.122L-37.647 71.122Q-37.486 71.145-37.486 71.289Q-37.486 71.361-37.532 71.407Q-37.578 71.453-37.647 71.463L-39.889 71.463L-39.889 73.719Q-39.913 73.880-40.056 73.880Q-40.128 73.880-40.174 73.834Q-40.220 73.788-40.231 73.719\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M208.677-37.926h31.298v-17.072h-31.298Z\"\u002F>\u003Cg transform=\"translate(260.99 -117.751)\">\u003Cpath d=\"M-42.815 71.528Q-42.815 71.200-42.680 70.899Q-42.545 70.599-42.309 70.378Q-42.073 70.158-41.769 70.038Q-41.464 69.918-41.140 69.918Q-40.634 69.918-40.285 70.021Q-39.937 70.123-39.937 70.499Q-39.937 70.646-40.034 70.747Q-40.131 70.848-40.278 70.848Q-40.432 70.848-40.531 70.749Q-40.630 70.650-40.630 70.499Q-40.630 70.311-40.490 70.219Q-40.692 70.168-41.133 70.168Q-41.488 70.168-41.717 70.364Q-41.946 70.561-42.047 70.870Q-42.148 71.180-42.148 71.528Q-42.148 71.877-42.022 72.183Q-41.895 72.489-41.640 72.673Q-41.386 72.858-41.030 72.858Q-40.808 72.858-40.624 72.774Q-40.439 72.690-40.304 72.535Q-40.169 72.379-40.111 72.171Q-40.097 72.116-40.043 72.116L-39.930 72.116Q-39.899 72.116-39.877 72.140Q-39.855 72.164-39.855 72.198L-39.855 72.219Q-39.940 72.506-40.128 72.704Q-40.316 72.902-40.581 73.005Q-40.846 73.107-41.140 73.107Q-41.570 73.107-41.958 72.901Q-42.346 72.694-42.580 72.331Q-42.815 71.969-42.815 71.528M-37.137 74.789Q-37.688 74.389-38.058 73.834Q-38.429 73.278-38.610 72.632Q-38.792 71.986-38.792 71.289Q-38.792 70.776-38.691 70.281Q-38.590 69.785-38.385 69.334Q-38.180 68.883-37.867 68.491Q-37.554 68.100-37.137 67.796Q-37.127 67.792-37.120 67.791Q-37.113 67.789-37.103 67.789L-37.035 67.789Q-37.001 67.789-36.978 67.813Q-36.956 67.837-36.956 67.874Q-36.956 67.919-36.983 67.936Q-37.332 68.237-37.585 68.621Q-37.838 69.006-37.990 69.447Q-38.142 69.888-38.214 70.344Q-38.286 70.800-38.286 71.289Q-38.286 72.290-37.976 73.177Q-37.667 74.064-36.983 74.649Q-36.956 74.666-36.956 74.710Q-36.956 74.748-36.978 74.772Q-37.001 74.796-37.035 74.796L-37.103 74.796Q-37.110 74.792-37.119 74.791Q-37.127 74.789-37.137 74.789M-35.620 72.198L-35.620 70.301L-36.259 70.301L-36.259 70.079Q-35.941 70.079-35.724 69.869Q-35.507 69.659-35.406 69.349Q-35.305 69.040-35.305 68.732L-35.039 68.732L-35.039 70.021L-33.962 70.021L-33.962 70.301L-35.039 70.301L-35.039 72.185Q-35.039 72.461-34.934 72.660Q-34.830 72.858-34.570 72.858Q-34.413 72.858-34.307 72.754Q-34.201 72.649-34.152 72.496Q-34.102 72.342-34.102 72.185L-34.102 71.771L-33.836 71.771L-33.836 72.198Q-33.836 72.424-33.935 72.634Q-34.034 72.844-34.218 72.976Q-34.403 73.107-34.632 73.107Q-35.069 73.107-35.345 72.870Q-35.620 72.632-35.620 72.198M-32.704 74.796L-32.773 74.796Q-32.807 74.796-32.829 74.770Q-32.851 74.745-32.851 74.710Q-32.851 74.666-32.820 74.649Q-32.465 74.345-32.215 73.955Q-31.966 73.565-31.814 73.133Q-31.662 72.701-31.592 72.232Q-31.522 71.764-31.522 71.289Q-31.522 70.810-31.592 70.344Q-31.662 69.877-31.816 69.442Q-31.969 69.006-32.221 68.618Q-32.472 68.230-32.820 67.936Q-32.851 67.919-32.851 67.874Q-32.851 67.840-32.829 67.815Q-32.807 67.789-32.773 67.789L-32.704 67.789Q-32.694 67.789-32.685 67.791Q-32.677 67.792-32.667 67.796Q-32.123 68.196-31.751 68.749Q-31.378 69.303-31.197 69.949Q-31.016 70.595-31.016 71.289Q-31.016 71.990-31.197 72.637Q-31.378 73.285-31.752 73.839Q-32.127 74.393-32.667 74.789Q-32.677 74.789-32.685 74.791Q-32.694 74.792-32.704 74.796\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-27.28-46.462h63.864\"\u002F>\u003Cpath stroke=\"none\" d=\"m39.183-46.462-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M56.655-46.462h70.977\"\u002F>\u003Cpath stroke=\"none\" d=\"m130.232-46.462-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M147.704-46.462h57.973\"\u002F>\u003Cpath stroke=\"none\" d=\"m208.277-46.462-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M57.878 4.753c0-5.5-4.459-9.959-9.959-9.959s-9.958 4.459-9.958 9.959 4.458 9.958 9.958 9.958 9.959-4.458 9.959-9.958Zm-9.959 0\"\u002F>\u003Cg transform=\"translate(89.81 -65.856)\">\u003Cpath d=\"M-41.017 73.039L-42.750 73.039L-42.750 72.759Q-42.524 72.759-42.375 72.725Q-42.227 72.690-42.227 72.550L-42.227 70.301L-42.815 70.301L-42.815 70.021L-42.227 70.021L-42.227 69.204Q-42.227 68.886-42.049 68.638Q-41.871 68.391-41.581 68.250Q-41.290 68.110-40.979 68.110Q-40.723 68.110-40.519 68.252Q-40.316 68.394-40.316 68.637Q-40.316 68.773-40.415 68.872Q-40.514 68.972-40.651 68.972Q-40.788 68.972-40.887 68.872Q-40.986 68.773-40.986 68.637Q-40.986 68.456-40.846 68.363Q-40.924 68.336-41.024 68.336Q-41.232 68.336-41.386 68.469Q-41.540 68.602-41.620 68.806Q-41.700 69.009-41.700 69.218L-41.700 70.021L-40.812 70.021L-40.812 70.301L-41.673 70.301L-41.673 72.550Q-41.673 72.759-41.017 72.759\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M109.093 44.586c0-5.5-4.459-9.958-9.959-9.958s-9.958 4.458-9.958 9.958 4.458 9.959 9.958 9.959 9.959-4.459 9.959-9.959Zm-9.959 0\"\u002F>\u003Cg transform=\"translate(140.27 -27.626)\">\u003Cpath d=\"M-42.856 73.572Q-42.856 73.326-42.659 73.142Q-42.462 72.957-42.206 72.878Q-42.343 72.766-42.415 72.605Q-42.486 72.444-42.486 72.263Q-42.486 71.942-42.275 71.696Q-42.609 71.398-42.609 70.988Q-42.609 70.527-42.220 70.240Q-41.830 69.953-41.352 69.953Q-40.880 69.953-40.545 70.199Q-40.371 70.045-40.160 69.963Q-39.950 69.881-39.721 69.881Q-39.557 69.881-39.436 69.988Q-39.315 70.096-39.315 70.260Q-39.315 70.356-39.386 70.428Q-39.458 70.499-39.550 70.499Q-39.650 70.499-39.720 70.426Q-39.790 70.352-39.790 70.253Q-39.790 70.199-39.776 70.168L-39.769 70.154Q-39.762 70.134-39.754 70.123Q-39.745 70.113-39.742 70.106Q-40.097 70.106-40.384 70.329Q-40.097 70.622-40.097 70.988Q-40.097 71.303-40.282 71.535Q-40.466 71.768-40.755 71.896Q-41.044 72.024-41.352 72.024Q-41.553 72.024-41.745 71.974Q-41.936 71.925-42.114 71.815Q-42.206 71.942-42.206 72.085Q-42.206 72.267-42.078 72.402Q-41.950 72.537-41.765 72.537L-41.133 72.537Q-40.685 72.537-40.316 72.608Q-39.947 72.680-39.687 72.909Q-39.427 73.138-39.427 73.572Q-39.427 73.893-39.723 74.095Q-40.019 74.297-40.422 74.386Q-40.825 74.475-41.140 74.475Q-41.458 74.475-41.861 74.386Q-42.264 74.297-42.560 74.095Q-42.856 73.893-42.856 73.572M-42.401 73.572Q-42.401 73.801-42.182 73.950Q-41.963 74.099-41.671 74.167Q-41.379 74.235-41.140 74.235Q-40.976 74.235-40.767 74.199Q-40.559 74.164-40.352 74.083Q-40.145 74.003-40.014 73.875Q-39.882 73.747-39.882 73.572Q-39.882 73.220-40.263 73.126Q-40.644 73.032-41.147 73.032L-41.765 73.032Q-42.004 73.032-42.203 73.183Q-42.401 73.333-42.401 73.572M-41.352 71.785Q-40.685 71.785-40.685 70.988Q-40.685 70.188-41.352 70.188Q-42.022 70.188-42.022 70.988Q-42.022 71.785-41.352 71.785\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M109.093 1.907c0-5.5-4.459-9.958-9.959-9.958s-9.958 4.458-9.958 9.958 4.458 9.959 9.958 9.959 9.959-4.459 9.959-9.959Zm-9.959 0\"\u002F>\u003Cg transform=\"translate(141.132 -68.778)\">\u003Cpath d=\"M-41.198 73.039L-42.750 73.039L-42.750 72.759Q-42.524 72.759-42.375 72.725Q-42.227 72.690-42.227 72.550L-42.227 70.701Q-42.227 70.513-42.275 70.429Q-42.322 70.346-42.420 70.327Q-42.517 70.308-42.729 70.308L-42.729 70.028L-41.673 69.953L-41.673 72.550Q-41.673 72.690-41.541 72.725Q-41.410 72.759-41.198 72.759L-41.198 73.039M-42.469 68.732Q-42.469 68.561-42.346 68.442Q-42.223 68.322-42.052 68.322Q-41.885 68.322-41.762 68.442Q-41.639 68.561-41.639 68.732Q-41.639 68.907-41.762 69.030Q-41.885 69.153-42.052 69.153Q-42.223 69.153-42.346 69.030Q-42.469 68.907-42.469 68.732\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M147.504 44.586a8.536 8.536 0 1 0-17.072 0 8.536 8.536 0 0 0 17.072 0Zm-8.536 0\"\u002F>\u003Cg transform=\"translate(179.997 -26.946)\">\u003Cpath d=\"M-41.632 73.039L-42.955 73.039L-42.955 72.759Q-42.394 72.759-42.015 72.359L-41.300 71.562L-42.213 70.513Q-42.350 70.366-42.498 70.334Q-42.647 70.301-42.914 70.301L-42.914 70.021L-41.413 70.021L-41.413 70.301Q-41.605 70.301-41.605 70.435Q-41.605 70.465-41.574 70.513L-40.979 71.197L-40.538 70.701Q-40.425 70.571-40.425 70.455Q-40.425 70.393-40.463 70.347Q-40.501 70.301-40.559 70.301L-40.559 70.021L-39.243 70.021L-39.243 70.301Q-39.803 70.301-40.183 70.701L-40.805 71.402L-39.810 72.550Q-39.711 72.649-39.610 72.694Q-39.509 72.738-39.398 72.748Q-39.287 72.759-39.109 72.759L-39.109 73.039L-40.603 73.039L-40.603 72.759Q-40.538 72.759-40.478 72.725Q-40.419 72.690-40.419 72.625Q-40.419 72.578-40.449 72.550L-41.126 71.764L-41.659 72.359Q-41.772 72.489-41.772 72.605Q-41.772 72.670-41.731 72.714Q-41.690 72.759-41.632 72.759\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M234.285 44.586c0-5.5-4.459-9.958-9.959-9.958s-9.958 4.458-9.958 9.958 4.458 9.959 9.958 9.959 9.959-4.459 9.959-9.959Zm-9.959 0\"\u002F>\u003Cg transform=\"translate(265.463 -26.946)\">\u003Cpath d=\"M-42.856 71.556Q-42.856 71.214-42.721 70.915Q-42.586 70.616-42.346 70.392Q-42.107 70.168-41.789 70.043Q-41.471 69.918-41.140 69.918Q-40.695 69.918-40.296 70.134Q-39.896 70.349-39.661 70.727Q-39.427 71.104-39.427 71.556Q-39.427 71.897-39.569 72.181Q-39.711 72.465-39.955 72.672Q-40.200 72.878-40.509 72.993Q-40.818 73.107-41.140 73.107Q-41.570 73.107-41.972 72.906Q-42.374 72.704-42.615 72.352Q-42.856 72-42.856 71.556M-41.140 72.858Q-40.538 72.858-40.314 72.480Q-40.090 72.102-40.090 71.470Q-40.090 70.858-40.325 70.499Q-40.559 70.141-41.140 70.141Q-42.192 70.141-42.192 71.470Q-42.192 72.102-41.967 72.480Q-41.741 72.858-41.140 72.858\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M232.862-.938a8.536 8.536 0 1 0-17.071 0 8.536 8.536 0 0 0 17.071 0Zm-8.536 0\"\u002F>\u003Cg transform=\"translate(265.355 -72.47)\">\u003Cpath d=\"M-41.632 73.039L-42.955 73.039L-42.955 72.759Q-42.394 72.759-42.015 72.359L-41.300 71.562L-42.213 70.513Q-42.350 70.366-42.498 70.334Q-42.647 70.301-42.914 70.301L-42.914 70.021L-41.413 70.021L-41.413 70.301Q-41.605 70.301-41.605 70.435Q-41.605 70.465-41.574 70.513L-40.979 71.197L-40.538 70.701Q-40.425 70.571-40.425 70.455Q-40.425 70.393-40.463 70.347Q-40.501 70.301-40.559 70.301L-40.559 70.021L-39.243 70.021L-39.243 70.301Q-39.803 70.301-40.183 70.701L-40.805 71.402L-39.810 72.550Q-39.711 72.649-39.610 72.694Q-39.509 72.738-39.398 72.748Q-39.287 72.759-39.109 72.759L-39.109 73.039L-40.603 73.039L-40.603 72.759Q-40.538 72.759-40.478 72.725Q-40.419 72.690-40.419 72.625Q-40.419 72.578-40.449 72.550L-41.126 71.764L-41.659 72.359Q-41.772 72.489-41.772 72.605Q-41.772 72.670-41.731 72.714Q-41.690 72.759-41.632 72.759\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M168.843-13.742h25.608v-14.226h-25.608Z\"\u002F>\u003Cg transform=\"translate(216.804 -91.464)\">\u003Cpath d=\"M-42.288 72.198L-42.288 70.301L-42.927 70.301L-42.927 70.079Q-42.609 70.079-42.392 69.869Q-42.175 69.659-42.075 69.349Q-41.974 69.040-41.974 68.732L-41.707 68.732L-41.707 70.021L-40.630 70.021L-40.630 70.301L-41.707 70.301L-41.707 72.185Q-41.707 72.461-41.603 72.660Q-41.499 72.858-41.239 72.858Q-41.082 72.858-40.976 72.754Q-40.870 72.649-40.820 72.496Q-40.771 72.342-40.771 72.185L-40.771 71.771L-40.504 71.771L-40.504 72.198Q-40.504 72.424-40.603 72.634Q-40.702 72.844-40.887 72.976Q-41.071 73.107-41.300 73.107Q-41.738 73.107-42.013 72.870Q-42.288 72.632-42.288 72.198M-39.636 72.311Q-39.636 71.979-39.412 71.752Q-39.188 71.525-38.845 71.397Q-38.501 71.268-38.129 71.216Q-37.756 71.163-37.452 71.163L-37.452 70.910Q-37.452 70.705-37.559 70.525Q-37.667 70.346-37.848 70.243Q-38.029 70.141-38.238 70.141Q-38.645 70.141-38.880 70.233Q-38.792 70.270-38.745 70.354Q-38.699 70.438-38.699 70.540Q-38.699 70.636-38.745 70.715Q-38.792 70.793-38.872 70.838Q-38.952 70.882-39.041 70.882Q-39.191 70.882-39.292 70.785Q-39.393 70.687-39.393 70.540Q-39.393 69.918-38.238 69.918Q-38.026 69.918-37.776 69.982Q-37.527 70.045-37.325 70.164Q-37.124 70.284-36.997 70.469Q-36.871 70.653-36.871 70.896L-36.871 72.472Q-36.871 72.588-36.809 72.684Q-36.748 72.779-36.635 72.779Q-36.525 72.779-36.461 72.685Q-36.396 72.591-36.396 72.472L-36.396 72.024L-36.129 72.024L-36.129 72.472Q-36.129 72.742-36.356 72.907Q-36.584 73.073-36.864 73.073Q-37.072 73.073-37.209 72.919Q-37.346 72.766-37.370 72.550Q-37.517 72.817-37.799 72.962Q-38.081 73.107-38.405 73.107Q-38.682 73.107-38.966 73.032Q-39.250 72.957-39.443 72.778Q-39.636 72.598-39.636 72.311M-39.021 72.311Q-39.021 72.485-38.920 72.615Q-38.819 72.745-38.663 72.815Q-38.508 72.885-38.344 72.885Q-38.125 72.885-37.917 72.788Q-37.708 72.690-37.580 72.509Q-37.452 72.328-37.452 72.102L-37.452 71.374Q-37.776 71.374-38.142 71.465Q-38.508 71.556-38.764 71.768Q-39.021 71.979-39.021 72.311M-34.030 73.039L-35.664 73.039L-35.664 72.759Q-35.435 72.759-35.286 72.725Q-35.138 72.690-35.138 72.550L-35.138 70.701Q-35.138 70.431-35.245 70.370Q-35.353 70.308-35.664 70.308L-35.664 70.028L-34.605 69.953L-34.605 70.602Q-34.434 70.294-34.129 70.123Q-33.825 69.953-33.480 69.953Q-32.974 69.953-32.691 70.176Q-32.407 70.400-32.407 70.896L-32.407 72.550Q-32.407 72.687-32.258 72.723Q-32.109 72.759-31.884 72.759L-31.884 73.039L-33.514 73.039L-33.514 72.759Q-33.285 72.759-33.137 72.725Q-32.988 72.690-32.988 72.550L-32.988 70.910Q-32.988 70.575-33.108 70.375Q-33.227 70.175-33.542 70.175Q-33.812 70.175-34.046 70.311Q-34.280 70.448-34.418 70.682Q-34.557 70.916-34.557 71.190L-34.557 72.550Q-34.557 72.687-34.406 72.723Q-34.256 72.759-34.030 72.759L-34.030 73.039M-29.614 73.039L-31.248 73.039L-31.248 72.759Q-31.019 72.759-30.870 72.725Q-30.722 72.690-30.722 72.550L-30.722 68.931Q-30.722 68.661-30.829 68.599Q-30.937 68.538-31.248 68.538L-31.248 68.257L-30.168 68.182L-30.168 70.568Q-30.062 70.383-29.884 70.241Q-29.707 70.100-29.498 70.026Q-29.290 69.953-29.064 69.953Q-28.558 69.953-28.275 70.176Q-27.991 70.400-27.991 70.896L-27.991 72.550Q-27.991 72.687-27.842 72.723Q-27.693 72.759-27.468 72.759L-27.468 73.039L-29.098 73.039L-29.098 72.759Q-28.869 72.759-28.721 72.725Q-28.572 72.690-28.572 72.550L-28.572 70.910Q-28.572 70.575-28.691 70.375Q-28.811 70.175-29.126 70.175Q-29.396 70.175-29.630 70.311Q-29.864 70.448-30.002 70.682Q-30.141 70.916-30.141 71.190L-30.141 72.550Q-30.141 72.687-29.990 72.723Q-29.840 72.759-29.614 72.759\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M265.583 53.122h31.298V36.051h-31.298Z\"\u002F>\u003Cg transform=\"translate(317.465 -26.703)\">\u003Cpath d=\"M-41.133 73.039L-42.767 73.039L-42.767 72.759Q-42.538 72.759-42.389 72.725Q-42.240 72.690-42.240 72.550L-42.240 68.931Q-42.240 68.661-42.348 68.599Q-42.456 68.538-42.767 68.538L-42.767 68.257L-41.687 68.182L-41.687 70.568Q-41.581 70.383-41.403 70.241Q-41.225 70.100-41.017 70.026Q-40.808 69.953-40.583 69.953Q-40.077 69.953-39.793 70.176Q-39.509 70.400-39.509 70.896L-39.509 72.550Q-39.509 72.687-39.361 72.723Q-39.212 72.759-38.986 72.759L-38.986 73.039L-40.617 73.039L-40.617 72.759Q-40.388 72.759-40.239 72.725Q-40.090 72.690-40.090 72.550L-40.090 70.910Q-40.090 70.575-40.210 70.375Q-40.330 70.175-40.644 70.175Q-40.914 70.175-41.148 70.311Q-41.382 70.448-41.521 70.682Q-41.659 70.916-41.659 71.190L-41.659 72.550Q-41.659 72.687-41.509 72.723Q-41.358 72.759-41.133 72.759L-41.133 73.039M-36.269 74.789Q-36.819 74.389-37.190 73.834Q-37.561 73.278-37.742 72.632Q-37.923 71.986-37.923 71.289Q-37.923 70.776-37.823 70.281Q-37.722 69.785-37.517 69.334Q-37.312 68.883-36.999 68.491Q-36.686 68.100-36.269 67.796Q-36.259 67.792-36.252 67.791Q-36.245 67.789-36.235 67.789L-36.167 67.789Q-36.132 67.789-36.110 67.813Q-36.088 67.837-36.088 67.874Q-36.088 67.919-36.115 67.936Q-36.464 68.237-36.717 68.621Q-36.970 69.006-37.122 69.447Q-37.274 69.888-37.346 70.344Q-37.418 70.800-37.418 71.289Q-37.418 72.290-37.108 73.177Q-36.799 74.064-36.115 74.649Q-36.088 74.666-36.088 74.710Q-36.088 74.748-36.110 74.772Q-36.132 74.796-36.167 74.796L-36.235 74.796Q-36.242 74.792-36.250 74.791Q-36.259 74.789-36.269 74.789M-34.752 72.198L-34.752 70.301L-35.391 70.301L-35.391 70.079Q-35.073 70.079-34.856 69.869Q-34.639 69.659-34.538 69.349Q-34.437 69.040-34.437 68.732L-34.171 68.732L-34.171 70.021L-33.094 70.021L-33.094 70.301L-34.171 70.301L-34.171 72.185Q-34.171 72.461-34.066 72.660Q-33.962 72.858-33.702 72.858Q-33.545 72.858-33.439 72.754Q-33.333 72.649-33.284 72.496Q-33.234 72.342-33.234 72.185L-33.234 71.771L-32.967 71.771L-32.967 72.198Q-32.967 72.424-33.066 72.634Q-33.166 72.844-33.350 72.976Q-33.535 73.107-33.764 73.107Q-34.201 73.107-34.476 72.870Q-34.752 72.632-34.752 72.198M-31.836 74.796L-31.904 74.796Q-31.939 74.796-31.961 74.770Q-31.983 74.745-31.983 74.710Q-31.983 74.666-31.952 74.649Q-31.597 74.345-31.347 73.955Q-31.098 73.565-30.946 73.133Q-30.794 72.701-30.723 72.232Q-30.653 71.764-30.653 71.289Q-30.653 70.810-30.723 70.344Q-30.794 69.877-30.947 69.442Q-31.101 69.006-31.352 68.618Q-31.604 68.230-31.952 67.936Q-31.983 67.919-31.983 67.874Q-31.983 67.840-31.961 67.815Q-31.939 67.789-31.904 67.789L-31.836 67.789Q-31.826 67.789-31.817 67.791Q-31.809 67.792-31.798 67.796Q-31.255 68.196-30.882 68.749Q-30.510 69.303-30.329 69.949Q-30.148 70.595-30.148 71.289Q-30.148 71.990-30.329 72.637Q-30.510 73.285-30.884 73.839Q-31.258 74.393-31.798 74.789Q-31.809 74.789-31.817 74.791Q-31.826 74.792-31.836 74.796\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"m-27.28 11.804 63.095-5.917\"\u002F>\u003Cpath stroke=\"none\" d=\"M37.806 5.7 34.47 4.408l1.344 1.48-1.046 1.705\"\u002F>\u003Cpath fill=\"none\" d=\"M-27.28 61.658c23.984 0 46.747-28.453 73.517-45.665\"\u002F>\u003Cpath stroke=\"none\" d=\"m47.92 14.911-3.557.385 1.874.697-.144 1.994\"\u002F>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M47.92-5.406v-29.72\"\u002F>\u003Cpath stroke=\"none\" d=\"m47.92-37.726-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M109.293 44.586h18.34\"\u002F>\u003Cpath stroke=\"none\" d=\"m130.232 44.586-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M108.771 5.12c7.435 2.478 18.816 22.395 22.9 30.943\"\u002F>\u003Cpath stroke=\"none\" d=\"m132.79 38.41.085-4.65-1.205 2.303-2.549-.51\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M138.968 35.85v-70.977\"\u002F>\u003Cpath stroke=\"none\" d=\"m138.968-37.726-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M224.326-37.526c0 5.29-22.762 8.135-27.286 10.074\"\u002F>\u003Cpath stroke=\"none\" d=\"m194.65-26.427 4.644.272-2.254-1.297.615-2.526\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M194.65-15.076c12.605 5.602 29.676-.089 29.676 2.802\"\u002F>\u003Cpath stroke=\"none\" d=\"m224.326-9.674 2.08-4.16-2.08 1.56-2.08-1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M224.326 34.428v-24.03\"\u002F>\u003Cpath stroke=\"none\" d=\"m224.326 7.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M232.309 2.61c17.625 7.833 48.923 13.524 48.923 30.44\"\u002F>\u003Cpath stroke=\"none\" d=\"m281.232 35.65 2.08-4.16-2.08 1.56-2.08-1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-68.737 78.73v-150.8H306.84v150.8ZM306.84-72.07\" style=\"stroke-dasharray:3.0,3.0\"\u002F>\u003Cg transform=\"translate(315.574 .735)\">\u003Cpath d=\"M-37.821 73.039L-42.441 73.039Q-42.538 73.039-42.538 72.907Q-42.533 72.883-42.516 72.819Q-42.499 72.756-42.472 72.722Q-42.445 72.687-42.397 72.687Q-42.143 72.687-41.947 72.673Q-41.752 72.658-41.620 72.619Q-41.513 72.575-41.449 72.351L-40.077 66.857Q-40.058 66.760-40.058 66.711Q-40.058 66.560-40.819 66.560Q-40.917 66.560-40.917 66.428Q-40.888 66.306-40.863 66.257Q-40.839 66.208-40.751 66.208L-38.007 66.208Q-37.909 66.208-37.909 66.340Q-37.948 66.560-38.041 66.560Q-38.700 66.560-38.940 66.599Q-39.169 66.652-39.228 66.901L-40.600 72.390Q-40.639 72.541-40.649 72.619Q-40.649 72.687-40.341 72.687L-39.540 72.687Q-38.827 72.687-38.383 72.380Q-37.939 72.072-37.721 71.684Q-37.504 71.296-37.348 70.864Q-37.191 70.432-37.118 70.417L-36.967 70.417Q-36.869 70.417-36.869 70.549L-37.719 72.951Q-37.763 73.039-37.821 73.039M-35.971 73.259L-36.039 73.259Q-36.068 73.259-36.093 73.225Q-36.117 73.190-36.117 73.161L-35.541 70.827Q-35.522 70.768-35.458 70.768L-35.277 70.768Q-35.199 70.768-35.199 70.871Q-35.287 71.227-35.287 71.540Q-35.287 72.214-34.879 72.560Q-34.472 72.907-33.788 72.907Q-33.363 72.907-32.995 72.614Q-32.626 72.321-32.409 71.879Q-32.191 71.437-32.191 71.027Q-32.191 70.690-32.345 70.432Q-32.499 70.173-32.802 70.090L-33.900 69.768Q-34.355 69.636-34.611 69.255Q-34.867 68.874-34.867 68.391Q-34.867 67.810-34.540 67.246Q-34.213 66.682-33.676 66.335Q-33.139 65.988-32.567 65.988Q-32.152 65.988-31.801 66.154Q-31.449 66.320-31.269 66.657L-30.751 66.018Q-30.692 65.988-30.678 65.988L-30.619 65.988Q-30.580 65.988-30.556 66.022Q-30.531 66.057-30.531 66.091L-31.108 68.410Q-31.122 68.478-31.191 68.478L-31.371 68.478Q-31.459 68.478-31.459 68.371Q-31.391 67.961-31.391 67.697Q-31.391 67.053-31.676 66.682Q-31.962 66.310-32.602 66.310Q-32.904 66.310-33.183 66.452Q-33.461 66.594-33.683 66.848Q-33.905 67.102-34.027 67.397Q-34.150 67.692-34.150 67.990Q-34.150 68.312-33.996 68.557Q-33.842 68.801-33.549 68.879L-32.450 69.201Q-31.977 69.328-31.728 69.724Q-31.479 70.119-31.479 70.637Q-31.479 71.096-31.659 71.555Q-31.840 72.014-32.172 72.407Q-32.504 72.800-32.931 73.029Q-33.358 73.259-33.827 73.259Q-34.301 73.259-34.716 73.098Q-35.131 72.936-35.370 72.580L-35.898 73.229Q-35.956 73.259-35.971 73.259M-29.389 72.907Q-29.384 72.883-29.369 72.824Q-29.355 72.766-29.330 72.727Q-29.306 72.687-29.257 72.687Q-28.896 72.687-28.622 72.675Q-28.349 72.663-28.207 72.639Q-27.968 72.590-27.909 72.351L-26.537 66.857Q-26.498 66.701-26.498 66.628Q-26.498 66.560-26.801 66.560L-27.299 66.560Q-27.929 66.560-28.285 66.821Q-28.642 67.082-28.744 67.319Q-28.847 67.556-28.988 68.017Q-29.130 68.478-29.237 68.498L-29.389 68.498Q-29.491 68.498-29.491 68.371L-28.798 66.301Q-28.778 66.208-28.700 66.208L-23.261 66.208Q-23.158 66.208-23.158 66.340L-23.510 68.410Q-23.510 68.439-23.544 68.469Q-23.578 68.498-23.608 68.498L-23.759 68.498Q-23.861 68.498-23.861 68.371Q-23.730 67.663-23.730 67.287Q-23.730 66.848-23.983 66.704Q-24.237 66.560-24.721 66.560L-25.209 66.560Q-25.473 66.560-25.548 66.606Q-25.624 66.652-25.688 66.901L-27.060 72.390Q-27.079 72.468-27.079 72.487Q-27.079 72.595-27.001 72.619Q-26.845 72.658-26.562 72.673Q-26.278 72.687-25.932 72.687Q-25.829 72.687-25.829 72.819Q-25.863 72.951-25.885 72.995Q-25.907 73.039-26 73.039L-29.291 73.039Q-29.389 73.039-29.389 72.907M-21.332 73.039L-23.358 73.039Q-23.461 73.039-23.461 72.907Q-23.456 72.883-23.439 72.819Q-23.422 72.756-23.395 72.722Q-23.368 72.687-23.319 72.687Q-22.460 72.687-22.299 72.077L-21 66.857Q-20.981 66.760-20.981 66.711Q-20.981 66.560-21.742 66.560Q-21.840 66.560-21.840 66.428Q-21.835 66.403-21.818 66.340Q-21.801 66.276-21.774 66.242Q-21.747 66.208-21.703 66.208L-20.019 66.208Q-19.970 66.208-19.931 66.237Q-19.892 66.267-19.892 66.320L-19.340 71.979L-15.961 66.320Q-15.893 66.208-15.780 66.208L-14.150 66.208Q-14.052 66.208-14.052 66.340Q-14.091 66.560-14.179 66.560Q-14.662 66.560-14.865 66.599Q-15.067 66.638-15.131 66.901L-16.503 72.390Q-16.513 72.414-16.518 72.448Q-16.523 72.482-16.532 72.541Q-16.532 72.687-15.771 72.687Q-15.673 72.687-15.673 72.819Q-15.692 72.946-15.717 72.993Q-15.741 73.039-15.839 73.039L-18.261 73.039Q-18.358 73.039-18.358 72.907Q-18.354 72.883-18.337 72.819Q-18.319 72.756-18.293 72.722Q-18.266 72.687-18.222 72.687Q-17.743 72.687-17.553 72.648Q-17.333 72.609-17.270 72.351L-15.858 66.677L-19.589 72.927Q-19.667 73.039-19.770 73.039L-19.862 73.039Q-19.989 73.039-19.989 72.927L-20.590 66.769L-21.923 72.121Q-21.952 72.209-21.952 72.287Q-21.952 72.531-21.750 72.609Q-21.547 72.687-21.259 72.687Q-21.161 72.687-21.161 72.819Q-21.181 72.946-21.210 72.993Q-21.239 73.039-21.332 73.039\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.300\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fsvg>\u003Cfigcaption class=\"tikz-cap\">A single LSTM cell. The previous cell state \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6528em;vertical-align:-0.2083em;\">\u003C\u002Fspan>\u003Cspan class=\"mord\">\u003Cspan class=\"mord mathbf\">c\u003C\u002Fspan>\u003Cspan class=\"msupsub\">\u003Cspan class=\"vlist-t vlist-t2\">\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.3011em;\">\u003Cspan style=\"top:-2.55em;margin-left:0em;margin-right:0.05em;\">\u003Cspan class=\"pstrut\" style=\"height:2.7em;\">\u003C\u002Fspan>\u003Cspan class=\"sizing reset-size6 size3 mtight\">\u003Cspan class=\"mord mtight\">\u003Cspan class=\"mord mathnormal mtight\">t\u003C\u002Fspan>\u003Cspan class=\"mbin mtight\">−\u003C\u002Fspan>\u003Cspan class=\"mord mtight\">1\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-s\">​\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.2083em;\">\u003Cspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> flows along the top, multiplied by the forget gate \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.8889em;vertical-align:-0.1944em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathnormal\" style=\"margin-right:0.1076em;\">f\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> then added to the gated candidate (\u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.625em;vertical-align:-0.1944em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathnormal\" style=\"margin-right:0.0359em;\">g\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> selected by the input gate \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.6595em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathnormal\">i\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>); the output gate \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.4306em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathnormal\">o\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> reads \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:1em;vertical-align:-0.25em;\">\u003C\u002Fspan>\u003Cspan class=\"mop\">tanh\u003C\u002Fspan>\u003Cspan class=\"mopen\">(\u003C\u002Fspan>\u003Cspan class=\"mord\">\u003Cspan class=\"mord mathbf\">c\u003C\u002Fspan>\u003Cspan class=\"msupsub\">\u003Cspan class=\"vlist-t vlist-t2\">\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.2806em;\">\u003Cspan style=\"top:-2.55em;margin-left:0em;margin-right:0.05em;\">\u003Cspan class=\"pstrut\" style=\"height:2.7em;\">\u003C\u002Fspan>\u003Cspan class=\"sizing reset-size6 size3 mtight\">\u003Cspan class=\"mord mathnormal mtight\">t\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-s\">​\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.15em;\">\u003Cspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"mclose\">)\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan> into the new hidden state \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.8444em;vertical-align:-0.15em;\">\u003C\u002Fspan>\u003Cspan class=\"mord\">\u003Cspan class=\"mord mathbf\">h\u003C\u002Fspan>\u003Cspan class=\"msupsub\">\u003Cspan class=\"vlist-t vlist-t2\">\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.2806em;\">\u003Cspan style=\"top:-2.55em;margin-left:0em;margin-right:0.05em;\">\u003Cspan class=\"pstrut\" style=\"height:2.7em;\">\u003C\u002Fspan>\u003Cspan class=\"sizing reset-size6 size3 mtight\">\u003Cspan class=\"mord mathnormal mtight\">t\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-s\">​\u003C\u002Fspan>\u003C\u002Fspan>\u003Cspan class=\"vlist-r\">\u003Cspan class=\"vlist\" style=\"height:0.15em;\">\u003Cspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>.\u003C\u002Ffigcaption>","\u003Csvg style=\"width:100%;max-width:304.059px;height:auto\" xmlns=\"http:\u002F\u002Fwww.w3.org\u002F2000\u002Fsvg\" viewBox=\"-75 -75 228.044 152.932\">\u003Cg stroke=\"currentColor\" style=\"stroke-miterlimit:10;stroke-width:.4\">\u003Cpath fill=\"none\" d=\"M-65.403 74.462h22.762V54.545h-22.762Z\"\u002F>\u003Cg transform=\"translate(61.068 1.75)\">\u003Cpath d=\"M-120.812 64.503L-122.135 64.503L-122.135 64.223Q-121.574 64.223-121.195 63.823L-120.480 63.026L-121.393 61.977Q-121.530 61.830-121.678 61.798Q-121.827 61.765-122.094 61.765L-122.094 61.485L-120.593 61.485L-120.593 61.765Q-120.785 61.765-120.785 61.899Q-120.785 61.929-120.754 61.977L-120.159 62.661L-119.718 62.165Q-119.605 62.035-119.605 61.919Q-119.605 61.857-119.643 61.811Q-119.681 61.765-119.739 61.765L-119.739 61.485L-118.423 61.485L-118.423 61.765Q-118.983 61.765-119.363 62.165L-119.985 62.866L-118.990 64.014Q-118.891 64.113-118.790 64.158Q-118.689 64.202-118.578 64.212Q-118.467 64.223-118.289 64.223L-118.289 64.503L-119.783 64.503L-119.783 64.223Q-119.718 64.223-119.658 64.189Q-119.599 64.154-119.599 64.089Q-119.599 64.042-119.629 64.014L-120.306 63.228L-120.839 63.823Q-120.952 63.953-120.952 64.069Q-120.952 64.134-120.911 64.178Q-120.870 64.223-120.812 64.223L-120.812 64.503M-115.664 66.253Q-116.215 65.853-116.586 65.298Q-116.956 64.742-117.138 64.096Q-117.319 63.450-117.319 62.753Q-117.319 62.240-117.218 61.745Q-117.117 61.249-116.912 60.798Q-116.707 60.347-116.394 59.955Q-116.081 59.564-115.664 59.260Q-115.654 59.256-115.647 59.255Q-115.641 59.253-115.630 59.253L-115.562 59.253Q-115.528 59.253-115.506 59.277Q-115.483 59.301-115.483 59.338Q-115.483 59.383-115.511 59.400Q-115.859 59.701-116.112 60.085Q-116.365 60.470-116.517 60.911Q-116.669 61.352-116.741 61.808Q-116.813 62.264-116.813 62.753Q-116.813 63.754-116.504 64.641Q-116.194 65.528-115.511 66.113Q-115.483 66.130-115.483 66.174Q-115.483 66.212-115.506 66.236Q-115.528 66.260-115.562 66.260L-115.630 66.260Q-115.637 66.256-115.646 66.255Q-115.654 66.253-115.664 66.253M-111.662 64.503L-114.191 64.503L-114.191 64.223Q-113.224 64.223-113.224 64.014L-113.224 60.395Q-113.617 60.583-114.239 60.583L-114.239 60.302Q-113.822 60.302-113.458 60.201Q-113.094 60.101-112.838 59.855L-112.711 59.855Q-112.646 59.872-112.629 59.940L-112.629 64.014Q-112.629 64.223-111.662 64.223L-111.662 64.503M-110.370 66.260L-110.438 66.260Q-110.473 66.260-110.495 66.234Q-110.517 66.209-110.517 66.174Q-110.517 66.130-110.486 66.113Q-110.131 65.809-109.881 65.419Q-109.632 65.029-109.480 64.597Q-109.328 64.165-109.257 63.696Q-109.187 63.228-109.187 62.753Q-109.187 62.274-109.257 61.808Q-109.328 61.341-109.481 60.906Q-109.635 60.470-109.886 60.082Q-110.138 59.694-110.486 59.400Q-110.517 59.383-110.517 59.338Q-110.517 59.304-110.495 59.279Q-110.473 59.253-110.438 59.253L-110.370 59.253Q-110.360 59.253-110.351 59.255Q-110.343 59.256-110.332 59.260Q-109.789 59.660-109.416 60.213Q-109.044 60.767-108.863 61.413Q-108.682 62.059-108.682 62.753Q-108.682 63.454-108.863 64.101Q-109.044 64.749-109.418 65.303Q-109.792 65.857-110.332 66.253Q-110.343 66.253-110.351 66.255Q-110.360 66.256-110.370 66.260\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-65.403 31.782h22.762V11.866h-22.762Z\"\u002F>\u003Cg transform=\"translate(67.047 -40.249)\">\u003Cpath d=\"M-120.197 64.503L-121.930 64.503L-121.930 64.223Q-121.704 64.223-121.555 64.189Q-121.407 64.154-121.407 64.014L-121.407 61.765L-121.995 61.765L-121.995 61.485L-121.407 61.485L-121.407 60.668Q-121.407 60.350-121.229 60.102Q-121.051 59.855-120.761 59.714Q-120.470 59.574-120.159 59.574Q-119.903 59.574-119.699 59.716Q-119.496 59.858-119.496 60.101Q-119.496 60.237-119.595 60.336Q-119.694 60.436-119.831 60.436Q-119.968 60.436-120.067 60.336Q-120.166 60.237-120.166 60.101Q-120.166 59.920-120.026 59.827Q-120.104 59.800-120.204 59.800Q-120.412 59.800-120.566 59.933Q-120.720 60.066-120.800 60.270Q-120.880 60.473-120.880 60.682L-120.880 61.485L-119.992 61.485L-119.992 61.765L-120.853 61.765L-120.853 64.014Q-120.853 64.223-120.197 64.223\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-65.403-10.897h22.762v-19.917h-22.762Z\"\u002F>\u003Cg transform=\"translate(66.078 -82.928)\">\u003Cpath d=\"M-121.188 64.503L-121.455 64.503L-121.455 60.395Q-121.455 60.125-121.562 60.063Q-121.670 60.002-121.981 60.002L-121.981 59.721L-120.901 59.646L-120.901 61.816Q-120.692 61.625-120.407 61.521Q-120.121 61.417-119.824 61.417Q-119.506 61.417-119.209 61.538Q-118.912 61.659-118.689 61.875Q-118.467 62.090-118.341 62.375Q-118.214 62.661-118.214 62.992Q-118.214 63.437-118.454 63.801Q-118.693 64.165-119.086 64.368Q-119.479 64.571-119.923 64.571Q-120.118 64.571-120.308 64.515Q-120.497 64.459-120.658 64.354Q-120.819 64.250-120.959 64.089L-121.188 64.503M-120.873 62.158L-120.873 63.775Q-120.737 64.035-120.496 64.192Q-120.255 64.349-119.978 64.349Q-119.684 64.349-119.472 64.242Q-119.260 64.134-119.127 63.942Q-118.994 63.751-118.935 63.512Q-118.877 63.273-118.877 62.992Q-118.877 62.633-118.971 62.329Q-119.065 62.025-119.293 61.832Q-119.520 61.639-119.886 61.639Q-120.186 61.639-120.453 61.775Q-120.720 61.912-120.873 62.158\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-65.403-54.998h22.762V-72.07h-22.762Z\"\u002F>\u003Cg transform=\"translate(61.068 -126.287)\">\u003Cpath d=\"M-121.660 65.638Q-121.530 65.706-121.393 65.706Q-121.222 65.706-121.072 65.617Q-120.921 65.528-120.810 65.383Q-120.699 65.238-120.621 65.070L-120.357 64.503L-121.526 61.977Q-121.601 61.830-121.731 61.798Q-121.861 61.765-122.094 61.765L-122.094 61.485L-120.573 61.485L-120.573 61.765Q-120.921 61.765-120.921 61.912Q-120.918 61.933-120.916 61.950Q-120.914 61.967-120.914 61.977L-120.057 63.836L-119.284 62.165Q-119.250 62.097-119.250 62.018Q-119.250 61.905-119.334 61.835Q-119.417 61.765-119.530 61.765L-119.530 61.485L-118.334 61.485L-118.334 61.765Q-118.553 61.765-118.725 61.869Q-118.898 61.974-118.990 62.165L-120.327 65.070Q-120.497 65.440-120.767 65.686Q-121.038 65.932-121.393 65.932Q-121.663 65.932-121.882 65.766Q-122.101 65.600-122.101 65.337Q-122.101 65.200-122.008 65.111Q-121.916 65.023-121.776 65.023Q-121.639 65.023-121.550 65.111Q-121.461 65.200-121.461 65.337Q-121.461 65.440-121.514 65.518Q-121.567 65.597-121.660 65.638M-115.664 66.253Q-116.215 65.853-116.586 65.298Q-116.956 64.742-117.138 64.096Q-117.319 63.450-117.319 62.753Q-117.319 62.240-117.218 61.745Q-117.117 61.249-116.912 60.798Q-116.707 60.347-116.394 59.955Q-116.081 59.564-115.664 59.260Q-115.654 59.256-115.647 59.255Q-115.641 59.253-115.630 59.253L-115.562 59.253Q-115.528 59.253-115.506 59.277Q-115.483 59.301-115.483 59.338Q-115.483 59.383-115.511 59.400Q-115.859 59.701-116.112 60.085Q-116.365 60.470-116.517 60.911Q-116.669 61.352-116.741 61.808Q-116.813 62.264-116.813 62.753Q-116.813 63.754-116.504 64.641Q-116.194 65.528-115.511 66.113Q-115.483 66.130-115.483 66.174Q-115.483 66.212-115.506 66.236Q-115.528 66.260-115.562 66.260L-115.630 66.260Q-115.637 66.256-115.646 66.255Q-115.654 66.253-115.664 66.253M-111.662 64.503L-114.191 64.503L-114.191 64.223Q-113.224 64.223-113.224 64.014L-113.224 60.395Q-113.617 60.583-114.239 60.583L-114.239 60.302Q-113.822 60.302-113.458 60.201Q-113.094 60.101-112.838 59.855L-112.711 59.855Q-112.646 59.872-112.629 59.940L-112.629 64.014Q-112.629 64.223-111.662 64.223L-111.662 64.503M-110.370 66.260L-110.438 66.260Q-110.473 66.260-110.495 66.234Q-110.517 66.209-110.517 66.174Q-110.517 66.130-110.486 66.113Q-110.131 65.809-109.881 65.419Q-109.632 65.029-109.480 64.597Q-109.328 64.165-109.257 63.696Q-109.187 63.228-109.187 62.753Q-109.187 62.274-109.257 61.808Q-109.328 61.341-109.481 60.906Q-109.635 60.470-109.886 60.082Q-110.138 59.694-110.486 59.400Q-110.517 59.383-110.517 59.338Q-110.517 59.304-110.495 59.279Q-110.473 59.253-110.438 59.253L-110.370 59.253Q-110.360 59.253-110.351 59.255Q-110.343 59.256-110.332 59.260Q-109.789 59.660-109.416 60.213Q-109.044 60.767-108.863 61.413Q-108.682 62.059-108.682 62.753Q-108.682 63.454-108.863 64.101Q-109.044 64.749-109.418 65.303Q-109.792 65.857-110.332 66.253Q-110.343 66.253-110.351 66.255Q-110.360 66.256-110.370 66.260\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-54.022 54.345V34.582\"\u002F>\u003Cpath stroke=\"none\" d=\"m-54.022 31.982-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-54.022 11.666v-63.864\"\u002F>\u003Cpath stroke=\"none\" d=\"m-54.022-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-54.022-31.014v-21.184\"\u002F>\u003Cpath stroke=\"none\" d=\"m-54.022-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M2.883 74.462h22.762V54.545H2.883Z\"\u002F>\u003Cg transform=\"translate(129.354 1.75)\">\u003Cpath d=\"M-120.812 64.503L-122.135 64.503L-122.135 64.223Q-121.574 64.223-121.195 63.823L-120.480 63.026L-121.393 61.977Q-121.530 61.830-121.678 61.798Q-121.827 61.765-122.094 61.765L-122.094 61.485L-120.593 61.485L-120.593 61.765Q-120.785 61.765-120.785 61.899Q-120.785 61.929-120.754 61.977L-120.159 62.661L-119.718 62.165Q-119.605 62.035-119.605 61.919Q-119.605 61.857-119.643 61.811Q-119.681 61.765-119.739 61.765L-119.739 61.485L-118.423 61.485L-118.423 61.765Q-118.983 61.765-119.363 62.165L-119.985 62.866L-118.990 64.014Q-118.891 64.113-118.790 64.158Q-118.689 64.202-118.578 64.212Q-118.467 64.223-118.289 64.223L-118.289 64.503L-119.783 64.503L-119.783 64.223Q-119.718 64.223-119.658 64.189Q-119.599 64.154-119.599 64.089Q-119.599 64.042-119.629 64.014L-120.306 63.228L-120.839 63.823Q-120.952 63.953-120.952 64.069Q-120.952 64.134-120.911 64.178Q-120.870 64.223-120.812 64.223L-120.812 64.503M-115.664 66.253Q-116.215 65.853-116.586 65.298Q-116.956 64.742-117.138 64.096Q-117.319 63.450-117.319 62.753Q-117.319 62.240-117.218 61.745Q-117.117 61.249-116.912 60.798Q-116.707 60.347-116.394 59.955Q-116.081 59.564-115.664 59.260Q-115.654 59.256-115.647 59.255Q-115.641 59.253-115.630 59.253L-115.562 59.253Q-115.528 59.253-115.506 59.277Q-115.483 59.301-115.483 59.338Q-115.483 59.383-115.511 59.400Q-115.859 59.701-116.112 60.085Q-116.365 60.470-116.517 60.911Q-116.669 61.352-116.741 61.808Q-116.813 62.264-116.813 62.753Q-116.813 63.754-116.504 64.641Q-116.194 65.528-115.511 66.113Q-115.483 66.130-115.483 66.174Q-115.483 66.212-115.506 66.236Q-115.528 66.260-115.562 66.260L-115.630 66.260Q-115.637 66.256-115.646 66.255Q-115.654 66.253-115.664 66.253M-111.662 64.503L-114.547 64.503L-114.547 64.301Q-114.547 64.271-114.519 64.243L-113.272 63.026Q-113.200 62.951-113.157 62.909Q-113.115 62.866-113.036 62.787Q-112.622 62.374-112.392 62.016Q-112.161 61.659-112.161 61.235Q-112.161 61.003-112.240 60.800Q-112.318 60.596-112.460 60.446Q-112.602 60.295-112.797 60.215Q-112.992 60.135-113.224 60.135Q-113.535 60.135-113.793 60.294Q-114.051 60.453-114.181 60.730L-114.161 60.730Q-113.993 60.730-113.885 60.841Q-113.778 60.952-113.778 61.116Q-113.778 61.273-113.887 61.386Q-113.996 61.499-114.161 61.499Q-114.321 61.499-114.434 61.386Q-114.547 61.273-114.547 61.116Q-114.547 60.740-114.338 60.453Q-114.130 60.166-113.795 60.010Q-113.460 59.855-113.104 59.855Q-112.681 59.855-112.301 60.013Q-111.922 60.172-111.688 60.489Q-111.454 60.805-111.454 61.235Q-111.454 61.546-111.594 61.815Q-111.734 62.083-111.939 62.288Q-112.144 62.493-112.506 62.775Q-112.869 63.057-112.978 63.153L-113.832 63.881L-113.190 63.881Q-112.927 63.881-112.638 63.879Q-112.349 63.878-112.130 63.869Q-111.912 63.860-111.894 63.843Q-111.833 63.778-111.795 63.611Q-111.758 63.443-111.720 63.201L-111.454 63.201L-111.662 64.503M-110.370 66.260L-110.438 66.260Q-110.473 66.260-110.495 66.234Q-110.517 66.209-110.517 66.174Q-110.517 66.130-110.486 66.113Q-110.131 65.809-109.881 65.419Q-109.632 65.029-109.480 64.597Q-109.328 64.165-109.257 63.696Q-109.187 63.228-109.187 62.753Q-109.187 62.274-109.257 61.808Q-109.328 61.341-109.481 60.906Q-109.635 60.470-109.886 60.082Q-110.138 59.694-110.486 59.400Q-110.517 59.383-110.517 59.338Q-110.517 59.304-110.495 59.279Q-110.473 59.253-110.438 59.253L-110.370 59.253Q-110.360 59.253-110.351 59.255Q-110.343 59.256-110.332 59.260Q-109.789 59.660-109.416 60.213Q-109.044 60.767-108.863 61.413Q-108.682 62.059-108.682 62.753Q-108.682 63.454-108.863 64.101Q-109.044 64.749-109.418 65.303Q-109.792 65.857-110.332 66.253Q-110.343 66.253-110.351 66.255Q-110.360 66.256-110.370 66.260\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M2.883 31.782h22.762V11.866H2.883Z\"\u002F>\u003Cg transform=\"translate(135.333 -40.249)\">\u003Cpath d=\"M-120.197 64.503L-121.930 64.503L-121.930 64.223Q-121.704 64.223-121.555 64.189Q-121.407 64.154-121.407 64.014L-121.407 61.765L-121.995 61.765L-121.995 61.485L-121.407 61.485L-121.407 60.668Q-121.407 60.350-121.229 60.102Q-121.051 59.855-120.761 59.714Q-120.470 59.574-120.159 59.574Q-119.903 59.574-119.699 59.716Q-119.496 59.858-119.496 60.101Q-119.496 60.237-119.595 60.336Q-119.694 60.436-119.831 60.436Q-119.968 60.436-120.067 60.336Q-120.166 60.237-120.166 60.101Q-120.166 59.920-120.026 59.827Q-120.104 59.800-120.204 59.800Q-120.412 59.800-120.566 59.933Q-120.720 60.066-120.800 60.270Q-120.880 60.473-120.880 60.682L-120.880 61.485L-119.992 61.485L-119.992 61.765L-120.853 61.765L-120.853 64.014Q-120.853 64.223-120.197 64.223\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M2.883-10.897h22.762v-19.917H2.883Z\"\u002F>\u003Cg transform=\"translate(134.364 -82.928)\">\u003Cpath d=\"M-121.188 64.503L-121.455 64.503L-121.455 60.395Q-121.455 60.125-121.562 60.063Q-121.670 60.002-121.981 60.002L-121.981 59.721L-120.901 59.646L-120.901 61.816Q-120.692 61.625-120.407 61.521Q-120.121 61.417-119.824 61.417Q-119.506 61.417-119.209 61.538Q-118.912 61.659-118.689 61.875Q-118.467 62.090-118.341 62.375Q-118.214 62.661-118.214 62.992Q-118.214 63.437-118.454 63.801Q-118.693 64.165-119.086 64.368Q-119.479 64.571-119.923 64.571Q-120.118 64.571-120.308 64.515Q-120.497 64.459-120.658 64.354Q-120.819 64.250-120.959 64.089L-121.188 64.503M-120.873 62.158L-120.873 63.775Q-120.737 64.035-120.496 64.192Q-120.255 64.349-119.978 64.349Q-119.684 64.349-119.472 64.242Q-119.260 64.134-119.127 63.942Q-118.994 63.751-118.935 63.512Q-118.877 63.273-118.877 62.992Q-118.877 62.633-118.971 62.329Q-119.065 62.025-119.293 61.832Q-119.520 61.639-119.886 61.639Q-120.186 61.639-120.453 61.775Q-120.720 61.912-120.873 62.158\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M2.883-54.998h22.762V-72.07H2.883Z\"\u002F>\u003Cg transform=\"translate(129.354 -126.287)\">\u003Cpath d=\"M-121.660 65.638Q-121.530 65.706-121.393 65.706Q-121.222 65.706-121.072 65.617Q-120.921 65.528-120.810 65.383Q-120.699 65.238-120.621 65.070L-120.357 64.503L-121.526 61.977Q-121.601 61.830-121.731 61.798Q-121.861 61.765-122.094 61.765L-122.094 61.485L-120.573 61.485L-120.573 61.765Q-120.921 61.765-120.921 61.912Q-120.918 61.933-120.916 61.950Q-120.914 61.967-120.914 61.977L-120.057 63.836L-119.284 62.165Q-119.250 62.097-119.250 62.018Q-119.250 61.905-119.334 61.835Q-119.417 61.765-119.530 61.765L-119.530 61.485L-118.334 61.485L-118.334 61.765Q-118.553 61.765-118.725 61.869Q-118.898 61.974-118.990 62.165L-120.327 65.070Q-120.497 65.440-120.767 65.686Q-121.038 65.932-121.393 65.932Q-121.663 65.932-121.882 65.766Q-122.101 65.600-122.101 65.337Q-122.101 65.200-122.008 65.111Q-121.916 65.023-121.776 65.023Q-121.639 65.023-121.550 65.111Q-121.461 65.200-121.461 65.337Q-121.461 65.440-121.514 65.518Q-121.567 65.597-121.660 65.638M-115.664 66.253Q-116.215 65.853-116.586 65.298Q-116.956 64.742-117.138 64.096Q-117.319 63.450-117.319 62.753Q-117.319 62.240-117.218 61.745Q-117.117 61.249-116.912 60.798Q-116.707 60.347-116.394 59.955Q-116.081 59.564-115.664 59.260Q-115.654 59.256-115.647 59.255Q-115.641 59.253-115.630 59.253L-115.562 59.253Q-115.528 59.253-115.506 59.277Q-115.483 59.301-115.483 59.338Q-115.483 59.383-115.511 59.400Q-115.859 59.701-116.112 60.085Q-116.365 60.470-116.517 60.911Q-116.669 61.352-116.741 61.808Q-116.813 62.264-116.813 62.753Q-116.813 63.754-116.504 64.641Q-116.194 65.528-115.511 66.113Q-115.483 66.130-115.483 66.174Q-115.483 66.212-115.506 66.236Q-115.528 66.260-115.562 66.260L-115.630 66.260Q-115.637 66.256-115.646 66.255Q-115.654 66.253-115.664 66.253M-111.662 64.503L-114.547 64.503L-114.547 64.301Q-114.547 64.271-114.519 64.243L-113.272 63.026Q-113.200 62.951-113.157 62.909Q-113.115 62.866-113.036 62.787Q-112.622 62.374-112.392 62.016Q-112.161 61.659-112.161 61.235Q-112.161 61.003-112.240 60.800Q-112.318 60.596-112.460 60.446Q-112.602 60.295-112.797 60.215Q-112.992 60.135-113.224 60.135Q-113.535 60.135-113.793 60.294Q-114.051 60.453-114.181 60.730L-114.161 60.730Q-113.993 60.730-113.885 60.841Q-113.778 60.952-113.778 61.116Q-113.778 61.273-113.887 61.386Q-113.996 61.499-114.161 61.499Q-114.321 61.499-114.434 61.386Q-114.547 61.273-114.547 61.116Q-114.547 60.740-114.338 60.453Q-114.130 60.166-113.795 60.010Q-113.460 59.855-113.104 59.855Q-112.681 59.855-112.301 60.013Q-111.922 60.172-111.688 60.489Q-111.454 60.805-111.454 61.235Q-111.454 61.546-111.594 61.815Q-111.734 62.083-111.939 62.288Q-112.144 62.493-112.506 62.775Q-112.869 63.057-112.978 63.153L-113.832 63.881L-113.190 63.881Q-112.927 63.881-112.638 63.879Q-112.349 63.878-112.130 63.869Q-111.912 63.860-111.894 63.843Q-111.833 63.778-111.795 63.611Q-111.758 63.443-111.720 63.201L-111.454 63.201L-111.662 64.503M-110.370 66.260L-110.438 66.260Q-110.473 66.260-110.495 66.234Q-110.517 66.209-110.517 66.174Q-110.517 66.130-110.486 66.113Q-110.131 65.809-109.881 65.419Q-109.632 65.029-109.480 64.597Q-109.328 64.165-109.257 63.696Q-109.187 63.228-109.187 62.753Q-109.187 62.274-109.257 61.808Q-109.328 61.341-109.481 60.906Q-109.635 60.470-109.886 60.082Q-110.138 59.694-110.486 59.400Q-110.517 59.383-110.517 59.338Q-110.517 59.304-110.495 59.279Q-110.473 59.253-110.438 59.253L-110.370 59.253Q-110.360 59.253-110.351 59.255Q-110.343 59.256-110.332 59.260Q-109.789 59.660-109.416 60.213Q-109.044 60.767-108.863 61.413Q-108.682 62.059-108.682 62.753Q-108.682 63.454-108.863 64.101Q-109.044 64.749-109.418 65.303Q-109.792 65.857-110.332 66.253Q-110.343 66.253-110.351 66.255Q-110.360 66.256-110.370 66.260\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M14.264 54.345V34.582\"\u002F>\u003Cpath stroke=\"none\" d=\"m14.264 31.982-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M14.264 11.666v-63.864\"\u002F>\u003Cpath stroke=\"none\" d=\"m14.264-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M14.264-31.014v-21.184\"\u002F>\u003Cpath stroke=\"none\" d=\"m14.264-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M71.17 74.462h22.762V54.545H71.169Z\"\u002F>\u003Cg transform=\"translate(197.64 1.75)\">\u003Cpath d=\"M-120.812 64.503L-122.135 64.503L-122.135 64.223Q-121.574 64.223-121.195 63.823L-120.480 63.026L-121.393 61.977Q-121.530 61.830-121.678 61.798Q-121.827 61.765-122.094 61.765L-122.094 61.485L-120.593 61.485L-120.593 61.765Q-120.785 61.765-120.785 61.899Q-120.785 61.929-120.754 61.977L-120.159 62.661L-119.718 62.165Q-119.605 62.035-119.605 61.919Q-119.605 61.857-119.643 61.811Q-119.681 61.765-119.739 61.765L-119.739 61.485L-118.423 61.485L-118.423 61.765Q-118.983 61.765-119.363 62.165L-119.985 62.866L-118.990 64.014Q-118.891 64.113-118.790 64.158Q-118.689 64.202-118.578 64.212Q-118.467 64.223-118.289 64.223L-118.289 64.503L-119.783 64.503L-119.783 64.223Q-119.718 64.223-119.658 64.189Q-119.599 64.154-119.599 64.089Q-119.599 64.042-119.629 64.014L-120.306 63.228L-120.839 63.823Q-120.952 63.953-120.952 64.069Q-120.952 64.134-120.911 64.178Q-120.870 64.223-120.812 64.223L-120.812 64.503M-115.664 66.253Q-116.215 65.853-116.586 65.298Q-116.956 64.742-117.138 64.096Q-117.319 63.450-117.319 62.753Q-117.319 62.240-117.218 61.745Q-117.117 61.249-116.912 60.798Q-116.707 60.347-116.394 59.955Q-116.081 59.564-115.664 59.260Q-115.654 59.256-115.647 59.255Q-115.641 59.253-115.630 59.253L-115.562 59.253Q-115.528 59.253-115.506 59.277Q-115.483 59.301-115.483 59.338Q-115.483 59.383-115.511 59.400Q-115.859 59.701-116.112 60.085Q-116.365 60.470-116.517 60.911Q-116.669 61.352-116.741 61.808Q-116.813 62.264-116.813 62.753Q-116.813 63.754-116.504 64.641Q-116.194 65.528-115.511 66.113Q-115.483 66.130-115.483 66.174Q-115.483 66.212-115.506 66.236Q-115.528 66.260-115.562 66.260L-115.630 66.260Q-115.637 66.256-115.646 66.255Q-115.654 66.253-115.664 66.253M-114.191 63.956Q-114.072 64.113-113.880 64.212Q-113.689 64.312-113.474 64.351Q-113.258 64.390-113.036 64.390Q-112.739 64.390-112.544 64.235Q-112.349 64.079-112.258 63.825Q-112.168 63.570-112.168 63.286Q-112.168 62.992-112.260 62.741Q-112.352 62.490-112.551 62.334Q-112.749 62.179-113.043 62.179L-113.559 62.179Q-113.586 62.179-113.612 62.153Q-113.638 62.128-113.638 62.104L-113.638 62.032Q-113.638 62.001-113.612 61.979Q-113.586 61.957-113.559 61.957L-113.118 61.926Q-112.756 61.926-112.535 61.569Q-112.315 61.211-112.315 60.822Q-112.315 60.494-112.510 60.290Q-112.705 60.087-113.036 60.087Q-113.323 60.087-113.576 60.171Q-113.829 60.254-113.993 60.442Q-113.846 60.442-113.745 60.557Q-113.644 60.671-113.644 60.822Q-113.644 60.972-113.750 61.082Q-113.856 61.191-114.014 61.191Q-114.174 61.191-114.284 61.082Q-114.393 60.972-114.393 60.822Q-114.393 60.497-114.184 60.278Q-113.976 60.060-113.660 59.957Q-113.344 59.855-113.036 59.855Q-112.718 59.855-112.390 59.959Q-112.062 60.063-111.835 60.285Q-111.607 60.507-111.607 60.822Q-111.607 61.256-111.894 61.581Q-112.182 61.905-112.616 62.052Q-112.305 62.117-112.024 62.283Q-111.744 62.449-111.566 62.707Q-111.389 62.965-111.389 63.286Q-111.389 63.696-111.633 64.006Q-111.877 64.315-112.258 64.479Q-112.640 64.643-113.036 64.643Q-113.405 64.643-113.762 64.530Q-114.120 64.418-114.364 64.168Q-114.608 63.919-114.608 63.549Q-114.608 63.378-114.492 63.266Q-114.376 63.153-114.205 63.153Q-114.096 63.153-114.005 63.204Q-113.914 63.255-113.860 63.348Q-113.805 63.440-113.805 63.549Q-113.805 63.717-113.918 63.836Q-114.031 63.956-114.191 63.956M-110.370 66.260L-110.438 66.260Q-110.473 66.260-110.495 66.234Q-110.517 66.209-110.517 66.174Q-110.517 66.130-110.486 66.113Q-110.131 65.809-109.881 65.419Q-109.632 65.029-109.480 64.597Q-109.328 64.165-109.257 63.696Q-109.187 63.228-109.187 62.753Q-109.187 62.274-109.257 61.808Q-109.328 61.341-109.481 60.906Q-109.635 60.470-109.886 60.082Q-110.138 59.694-110.486 59.400Q-110.517 59.383-110.517 59.338Q-110.517 59.304-110.495 59.279Q-110.473 59.253-110.438 59.253L-110.370 59.253Q-110.360 59.253-110.351 59.255Q-110.343 59.256-110.332 59.260Q-109.789 59.660-109.416 60.213Q-109.044 60.767-108.863 61.413Q-108.682 62.059-108.682 62.753Q-108.682 63.454-108.863 64.101Q-109.044 64.749-109.418 65.303Q-109.792 65.857-110.332 66.253Q-110.343 66.253-110.351 66.255Q-110.360 66.256-110.370 66.260\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M71.17 31.782h22.762V11.866H71.169Z\"\u002F>\u003Cg transform=\"translate(203.62 -40.249)\">\u003Cpath d=\"M-120.197 64.503L-121.930 64.503L-121.930 64.223Q-121.704 64.223-121.555 64.189Q-121.407 64.154-121.407 64.014L-121.407 61.765L-121.995 61.765L-121.995 61.485L-121.407 61.485L-121.407 60.668Q-121.407 60.350-121.229 60.102Q-121.051 59.855-120.761 59.714Q-120.470 59.574-120.159 59.574Q-119.903 59.574-119.699 59.716Q-119.496 59.858-119.496 60.101Q-119.496 60.237-119.595 60.336Q-119.694 60.436-119.831 60.436Q-119.968 60.436-120.067 60.336Q-120.166 60.237-120.166 60.101Q-120.166 59.920-120.026 59.827Q-120.104 59.800-120.204 59.800Q-120.412 59.800-120.566 59.933Q-120.720 60.066-120.800 60.270Q-120.880 60.473-120.880 60.682L-120.880 61.485L-119.992 61.485L-119.992 61.765L-120.853 61.765L-120.853 64.014Q-120.853 64.223-120.197 64.223\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M71.17-10.897h22.762v-19.917H71.169Z\"\u002F>\u003Cg transform=\"translate(202.65 -82.928)\">\u003Cpath d=\"M-121.188 64.503L-121.455 64.503L-121.455 60.395Q-121.455 60.125-121.562 60.063Q-121.670 60.002-121.981 60.002L-121.981 59.721L-120.901 59.646L-120.901 61.816Q-120.692 61.625-120.407 61.521Q-120.121 61.417-119.824 61.417Q-119.506 61.417-119.209 61.538Q-118.912 61.659-118.689 61.875Q-118.467 62.090-118.341 62.375Q-118.214 62.661-118.214 62.992Q-118.214 63.437-118.454 63.801Q-118.693 64.165-119.086 64.368Q-119.479 64.571-119.923 64.571Q-120.118 64.571-120.308 64.515Q-120.497 64.459-120.658 64.354Q-120.819 64.250-120.959 64.089L-121.188 64.503M-120.873 62.158L-120.873 63.775Q-120.737 64.035-120.496 64.192Q-120.255 64.349-119.978 64.349Q-119.684 64.349-119.472 64.242Q-119.260 64.134-119.127 63.942Q-118.994 63.751-118.935 63.512Q-118.877 63.273-118.877 62.992Q-118.877 62.633-118.971 62.329Q-119.065 62.025-119.293 61.832Q-119.520 61.639-119.886 61.639Q-120.186 61.639-120.453 61.775Q-120.720 61.912-120.873 62.158\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M71.17-54.998h22.762V-72.07H71.169Z\"\u002F>\u003Cg transform=\"translate(197.64 -126.287)\">\u003Cpath d=\"M-121.660 65.638Q-121.530 65.706-121.393 65.706Q-121.222 65.706-121.072 65.617Q-120.921 65.528-120.810 65.383Q-120.699 65.238-120.621 65.070L-120.357 64.503L-121.526 61.977Q-121.601 61.830-121.731 61.798Q-121.861 61.765-122.094 61.765L-122.094 61.485L-120.573 61.485L-120.573 61.765Q-120.921 61.765-120.921 61.912Q-120.918 61.933-120.916 61.950Q-120.914 61.967-120.914 61.977L-120.057 63.836L-119.284 62.165Q-119.250 62.097-119.250 62.018Q-119.250 61.905-119.334 61.835Q-119.417 61.765-119.530 61.765L-119.530 61.485L-118.334 61.485L-118.334 61.765Q-118.553 61.765-118.725 61.869Q-118.898 61.974-118.990 62.165L-120.327 65.070Q-120.497 65.440-120.767 65.686Q-121.038 65.932-121.393 65.932Q-121.663 65.932-121.882 65.766Q-122.101 65.600-122.101 65.337Q-122.101 65.200-122.008 65.111Q-121.916 65.023-121.776 65.023Q-121.639 65.023-121.550 65.111Q-121.461 65.200-121.461 65.337Q-121.461 65.440-121.514 65.518Q-121.567 65.597-121.660 65.638M-115.664 66.253Q-116.215 65.853-116.586 65.298Q-116.956 64.742-117.138 64.096Q-117.319 63.450-117.319 62.753Q-117.319 62.240-117.218 61.745Q-117.117 61.249-116.912 60.798Q-116.707 60.347-116.394 59.955Q-116.081 59.564-115.664 59.260Q-115.654 59.256-115.647 59.255Q-115.641 59.253-115.630 59.253L-115.562 59.253Q-115.528 59.253-115.506 59.277Q-115.483 59.301-115.483 59.338Q-115.483 59.383-115.511 59.400Q-115.859 59.701-116.112 60.085Q-116.365 60.470-116.517 60.911Q-116.669 61.352-116.741 61.808Q-116.813 62.264-116.813 62.753Q-116.813 63.754-116.504 64.641Q-116.194 65.528-115.511 66.113Q-115.483 66.130-115.483 66.174Q-115.483 66.212-115.506 66.236Q-115.528 66.260-115.562 66.260L-115.630 66.260Q-115.637 66.256-115.646 66.255Q-115.654 66.253-115.664 66.253M-114.191 63.956Q-114.072 64.113-113.880 64.212Q-113.689 64.312-113.474 64.351Q-113.258 64.390-113.036 64.390Q-112.739 64.390-112.544 64.235Q-112.349 64.079-112.258 63.825Q-112.168 63.570-112.168 63.286Q-112.168 62.992-112.260 62.741Q-112.352 62.490-112.551 62.334Q-112.749 62.179-113.043 62.179L-113.559 62.179Q-113.586 62.179-113.612 62.153Q-113.638 62.128-113.638 62.104L-113.638 62.032Q-113.638 62.001-113.612 61.979Q-113.586 61.957-113.559 61.957L-113.118 61.926Q-112.756 61.926-112.535 61.569Q-112.315 61.211-112.315 60.822Q-112.315 60.494-112.510 60.290Q-112.705 60.087-113.036 60.087Q-113.323 60.087-113.576 60.171Q-113.829 60.254-113.993 60.442Q-113.846 60.442-113.745 60.557Q-113.644 60.671-113.644 60.822Q-113.644 60.972-113.750 61.082Q-113.856 61.191-114.014 61.191Q-114.174 61.191-114.284 61.082Q-114.393 60.972-114.393 60.822Q-114.393 60.497-114.184 60.278Q-113.976 60.060-113.660 59.957Q-113.344 59.855-113.036 59.855Q-112.718 59.855-112.390 59.959Q-112.062 60.063-111.835 60.285Q-111.607 60.507-111.607 60.822Q-111.607 61.256-111.894 61.581Q-112.182 61.905-112.616 62.052Q-112.305 62.117-112.024 62.283Q-111.744 62.449-111.566 62.707Q-111.389 62.965-111.389 63.286Q-111.389 63.696-111.633 64.006Q-111.877 64.315-112.258 64.479Q-112.640 64.643-113.036 64.643Q-113.405 64.643-113.762 64.530Q-114.120 64.418-114.364 64.168Q-114.608 63.919-114.608 63.549Q-114.608 63.378-114.492 63.266Q-114.376 63.153-114.205 63.153Q-114.096 63.153-114.005 63.204Q-113.914 63.255-113.860 63.348Q-113.805 63.440-113.805 63.549Q-113.805 63.717-113.918 63.836Q-114.031 63.956-114.191 63.956M-110.370 66.260L-110.438 66.260Q-110.473 66.260-110.495 66.234Q-110.517 66.209-110.517 66.174Q-110.517 66.130-110.486 66.113Q-110.131 65.809-109.881 65.419Q-109.632 65.029-109.480 64.597Q-109.328 64.165-109.257 63.696Q-109.187 63.228-109.187 62.753Q-109.187 62.274-109.257 61.808Q-109.328 61.341-109.481 60.906Q-109.635 60.470-109.886 60.082Q-110.138 59.694-110.486 59.400Q-110.517 59.383-110.517 59.338Q-110.517 59.304-110.495 59.279Q-110.473 59.253-110.438 59.253L-110.370 59.253Q-110.360 59.253-110.351 59.255Q-110.343 59.256-110.332 59.260Q-109.789 59.660-109.416 60.213Q-109.044 60.767-108.863 61.413Q-108.682 62.059-108.682 62.753Q-108.682 63.454-108.863 64.101Q-109.044 64.749-109.418 65.303Q-109.792 65.857-110.332 66.253Q-110.343 66.253-110.351 66.255Q-110.360 66.256-110.370 66.260\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M82.55 54.345V34.582\"\u002F>\u003Cpath stroke=\"none\" d=\"m82.55 31.982-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M82.55 11.666v-63.864\"\u002F>\u003Cpath stroke=\"none\" d=\"m82.55-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M82.55-31.014v-21.184\"\u002F>\u003Cpath stroke=\"none\" d=\"m82.55-54.798-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-42.441 21.824H.083\"\u002F>\u003Cpath stroke=\"none\" d=\"m2.683 21.824-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M25.845 21.824H68.37\"\u002F>\u003Cpath stroke=\"none\" d=\"m70.97 21.824-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M70.97-20.855H28.444\"\u002F>\u003Cpath stroke=\"none\" d=\"m25.845-20.855 4.16 2.08-1.56-2.08 1.56-2.08\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M2.683-20.855H-39.84\"\u002F>\u003Cpath stroke=\"none\" d=\"m-42.441-20.855 4.16 2.08-1.56-2.08 1.56-2.08\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\">\u003Cg fill=\"var(--tk-accent)\" stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(218.35 -40.93)\">\u003Cpath d=\"M-120.197 64.503L-121.930 64.503L-121.930 64.223Q-121.704 64.223-121.555 64.189Q-121.407 64.154-121.407 64.014L-121.407 61.765L-121.995 61.765L-121.995 61.485L-121.407 61.485L-121.407 60.668Q-121.407 60.350-121.229 60.102Q-121.051 59.855-120.761 59.714Q-120.470 59.574-120.159 59.574Q-119.903 59.574-119.699 59.716Q-119.496 59.858-119.496 60.101Q-119.496 60.237-119.595 60.336Q-119.694 60.436-119.831 60.436Q-119.968 60.436-120.067 60.336Q-120.166 60.237-120.166 60.101Q-120.166 59.920-120.026 59.827Q-120.104 59.800-120.204 59.800Q-120.412 59.800-120.566 59.933Q-120.720 60.066-120.800 60.270Q-120.880 60.473-120.880 60.682L-120.880 61.485L-119.992 61.485L-119.992 61.765L-120.853 61.765L-120.853 64.014Q-120.853 64.223-120.197 64.223L-120.197 64.503M-119.558 63.020Q-119.558 62.678-119.423 62.379Q-119.288 62.080-119.048 61.856Q-118.809 61.632-118.491 61.507Q-118.173 61.382-117.842 61.382Q-117.397 61.382-116.997 61.598Q-116.598 61.813-116.363 62.191Q-116.129 62.568-116.129 63.020Q-116.129 63.361-116.271 63.645Q-116.413 63.929-116.657 64.136Q-116.902 64.342-117.211 64.457Q-117.520 64.571-117.842 64.571Q-118.272 64.571-118.674 64.370Q-119.076 64.168-119.317 63.816Q-119.558 63.464-119.558 63.020M-117.842 64.322Q-117.240 64.322-117.016 63.944Q-116.792 63.566-116.792 62.934Q-116.792 62.322-117.027 61.963Q-117.261 61.605-117.842 61.605Q-118.894 61.605-118.894 62.934Q-118.894 63.566-118.669 63.944Q-118.443 64.322-117.842 64.322M-113.785 64.503L-115.521 64.503L-115.521 64.223Q-115.292 64.223-115.143 64.189Q-114.995 64.154-114.995 64.014L-114.995 62.165Q-114.995 61.895-115.102 61.834Q-115.210 61.772-115.521 61.772L-115.521 61.492L-114.492 61.417L-114.492 62.124Q-114.362 61.816-114.120 61.617Q-113.877 61.417-113.559 61.417Q-113.340 61.417-113.169 61.541Q-112.998 61.666-112.998 61.878Q-112.998 62.015-113.098 62.114Q-113.197 62.213-113.330 62.213Q-113.467 62.213-113.566 62.114Q-113.665 62.015-113.665 61.878Q-113.665 61.738-113.566 61.639Q-113.856 61.639-114.056 61.835Q-114.256 62.032-114.349 62.326Q-114.441 62.620-114.441 62.900L-114.441 64.014Q-114.441 64.223-113.785 64.223L-113.785 64.503M-111.026 64.476L-112.007 61.977Q-112.069 61.834-112.187 61.799Q-112.305 61.765-112.520 61.765L-112.520 61.485L-111.040 61.485L-111.040 61.765Q-111.419 61.765-111.419 61.926Q-111.419 61.936-111.406 61.977L-110.691 63.809L-110.018 62.104Q-110.049 62.032-110.049 62.004Q-110.049 61.977-110.076 61.977Q-110.138 61.830-110.256 61.798Q-110.373 61.765-110.585 61.765L-110.585 61.485L-109.187 61.485L-109.187 61.765Q-109.563 61.765-109.563 61.926Q-109.563 61.957-109.557 61.977L-108.801 63.915L-108.114 62.165Q-108.094 62.114-108.094 62.059Q-108.094 61.919-108.206 61.842Q-108.319 61.765-108.459 61.765L-108.459 61.485L-107.239 61.485L-107.239 61.765Q-107.444 61.765-107.600 61.871Q-107.755 61.977-107.827 62.165L-108.733 64.476Q-108.767 64.571-108.880 64.571L-108.948 64.571Q-109.058 64.571-109.095 64.476L-109.878 62.473L-110.664 64.476Q-110.698 64.571-110.811 64.571L-110.879 64.571Q-110.989 64.571-111.026 64.476\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(218.35 -40.93)\">\u003Cpath d=\"M-106.852 63.775Q-106.852 63.443-106.629 63.216Q-106.405 62.989-106.061 62.861Q-105.718 62.732-105.345 62.680Q-104.973 62.627-104.668 62.627L-104.668 62.374Q-104.668 62.169-104.776 61.989Q-104.884 61.810-105.065 61.707Q-105.246 61.605-105.454 61.605Q-105.861 61.605-106.097 61.697Q-106.008 61.734-105.962 61.818Q-105.916 61.902-105.916 62.004Q-105.916 62.100-105.962 62.179Q-106.008 62.257-106.089 62.302Q-106.169 62.346-106.258 62.346Q-106.408 62.346-106.509 62.249Q-106.610 62.151-106.610 62.004Q-106.610 61.382-105.454 61.382Q-105.243 61.382-104.993 61.446Q-104.744 61.509-104.542 61.628Q-104.340 61.748-104.214 61.933Q-104.087 62.117-104.087 62.360L-104.087 63.936Q-104.087 64.052-104.026 64.148Q-103.964 64.243-103.851 64.243Q-103.742 64.243-103.677 64.149Q-103.612 64.055-103.612 63.936L-103.612 63.488L-103.346 63.488L-103.346 63.936Q-103.346 64.206-103.573 64.371Q-103.800 64.537-104.080 64.537Q-104.289 64.537-104.426 64.383Q-104.562 64.230-104.586 64.014Q-104.733 64.281-105.015 64.426Q-105.297 64.571-105.622 64.571Q-105.899 64.571-106.183 64.496Q-106.466 64.421-106.659 64.242Q-106.852 64.062-106.852 63.775M-106.237 63.775Q-106.237 63.949-106.136 64.079Q-106.036 64.209-105.880 64.279Q-105.725 64.349-105.560 64.349Q-105.342 64.349-105.133 64.252Q-104.925 64.154-104.797 63.973Q-104.668 63.792-104.668 63.566L-104.668 62.838Q-104.993 62.838-105.359 62.929Q-105.725 63.020-105.981 63.232Q-106.237 63.443-106.237 63.775M-101.179 64.503L-102.915 64.503L-102.915 64.223Q-102.686 64.223-102.537 64.189Q-102.389 64.154-102.389 64.014L-102.389 62.165Q-102.389 61.895-102.496 61.834Q-102.604 61.772-102.915 61.772L-102.915 61.492L-101.886 61.417L-101.886 62.124Q-101.756 61.816-101.514 61.617Q-101.271 61.417-100.953 61.417Q-100.734 61.417-100.563 61.541Q-100.392 61.666-100.392 61.878Q-100.392 62.015-100.492 62.114Q-100.591 62.213-100.724 62.213Q-100.861 62.213-100.960 62.114Q-101.059 62.015-101.059 61.878Q-101.059 61.738-100.960 61.639Q-101.250 61.639-101.450 61.835Q-101.650 62.032-101.743 62.326Q-101.835 62.620-101.835 62.900L-101.835 64.014Q-101.835 64.223-101.179 64.223L-101.179 64.503M-99.808 62.992Q-99.808 62.654-99.668 62.363Q-99.528 62.073-99.283 61.859Q-99.039 61.646-98.735 61.531Q-98.431 61.417-98.106 61.417Q-97.836 61.417-97.573 61.516Q-97.309 61.615-97.118 61.793L-97.118 60.395Q-97.118 60.125-97.226 60.063Q-97.333 60.002-97.644 60.002L-97.644 59.721L-96.568 59.646L-96.568 63.830Q-96.568 64.018-96.513 64.101Q-96.458 64.185-96.358 64.204Q-96.257 64.223-96.041 64.223L-96.041 64.503L-97.149 64.571L-97.149 64.154Q-97.566 64.571-98.191 64.571Q-98.622 64.571-98.995 64.359Q-99.367 64.148-99.588 63.787Q-99.808 63.426-99.808 62.992M-98.133 64.349Q-97.925 64.349-97.738 64.277Q-97.552 64.206-97.398 64.069Q-97.245 63.932-97.149 63.754L-97.149 62.145Q-97.234 61.998-97.380 61.878Q-97.525 61.758-97.694 61.699Q-97.863 61.639-98.044 61.639Q-98.605 61.639-98.873 62.028Q-99.142 62.418-99.142 62.999Q-99.142 63.570-98.907 63.960Q-98.673 64.349-98.133 64.349\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(218.35 -40.93)\">\u003Cpath d=\"M-91.045 65.860L-92.675 65.860L-92.675 65.580Q-92.446 65.580-92.297 65.545Q-92.149 65.511-92.149 65.371L-92.149 62.025Q-92.149 61.854-92.285 61.813Q-92.422 61.772-92.675 61.772L-92.675 61.492L-91.595 61.417L-91.595 61.823Q-91.373 61.622-91.086 61.519Q-90.798 61.417-90.491 61.417Q-90.064 61.417-89.700 61.630Q-89.336 61.844-89.122 62.208Q-88.908 62.572-88.908 62.992Q-88.908 63.437-89.148 63.801Q-89.387 64.165-89.780 64.368Q-90.173 64.571-90.617 64.571Q-90.884 64.571-91.132 64.471Q-91.379 64.370-91.567 64.189L-91.567 65.371Q-91.567 65.508-91.419 65.544Q-91.270 65.580-91.045 65.580L-91.045 65.860M-91.567 62.172L-91.567 63.782Q-91.434 64.035-91.191 64.192Q-90.949 64.349-90.672 64.349Q-90.344 64.349-90.091 64.148Q-89.838 63.946-89.705 63.628Q-89.571 63.310-89.571 62.992Q-89.571 62.763-89.636 62.534Q-89.701 62.305-89.829 62.107Q-89.958 61.909-90.152 61.789Q-90.347 61.670-90.580 61.670Q-90.874 61.670-91.142 61.799Q-91.410 61.929-91.567 62.172M-88.214 63.775Q-88.214 63.443-87.991 63.216Q-87.767 62.989-87.423 62.861Q-87.080 62.732-86.707 62.680Q-86.335 62.627-86.030 62.627L-86.030 62.374Q-86.030 62.169-86.138 61.989Q-86.246 61.810-86.427 61.707Q-86.608 61.605-86.816 61.605Q-87.223 61.605-87.459 61.697Q-87.370 61.734-87.324 61.818Q-87.278 61.902-87.278 62.004Q-87.278 62.100-87.324 62.179Q-87.370 62.257-87.451 62.302Q-87.531 62.346-87.620 62.346Q-87.770 62.346-87.871 62.249Q-87.972 62.151-87.972 62.004Q-87.972 61.382-86.816 61.382Q-86.605 61.382-86.355 61.446Q-86.106 61.509-85.904 61.628Q-85.702 61.748-85.576 61.933Q-85.449 62.117-85.449 62.360L-85.449 63.936Q-85.449 64.052-85.388 64.148Q-85.326 64.243-85.213 64.243Q-85.104 64.243-85.039 64.149Q-84.974 64.055-84.974 63.936L-84.974 63.488L-84.708 63.488L-84.708 63.936Q-84.708 64.206-84.935 64.371Q-85.162 64.537-85.442 64.537Q-85.651 64.537-85.788 64.383Q-85.924 64.230-85.948 64.014Q-86.095 64.281-86.377 64.426Q-86.659 64.571-86.984 64.571Q-87.261 64.571-87.545 64.496Q-87.828 64.421-88.021 64.242Q-88.214 64.062-88.214 63.775M-87.599 63.775Q-87.599 63.949-87.498 64.079Q-87.398 64.209-87.242 64.279Q-87.086 64.349-86.922 64.349Q-86.704 64.349-86.495 64.252Q-86.287 64.154-86.159 63.973Q-86.030 63.792-86.030 63.566L-86.030 62.838Q-86.355 62.838-86.721 62.929Q-87.086 63.020-87.343 63.232Q-87.599 63.443-87.599 63.775M-84.291 64.496L-84.291 63.433Q-84.291 63.409-84.263 63.382Q-84.236 63.355-84.212 63.355L-84.103 63.355Q-84.038 63.355-84.024 63.413Q-83.928 63.847-83.682 64.098Q-83.436 64.349-83.023 64.349Q-82.681 64.349-82.428 64.216Q-82.175 64.083-82.175 63.775Q-82.175 63.618-82.269 63.503Q-82.363 63.389-82.501 63.320Q-82.640 63.252-82.807 63.214L-83.388 63.115Q-83.744 63.047-84.017 62.826Q-84.291 62.606-84.291 62.264Q-84.291 62.015-84.180 61.840Q-84.068 61.666-83.882 61.567Q-83.696 61.468-83.481 61.425Q-83.265 61.382-83.023 61.382Q-82.609 61.382-82.329 61.564L-82.113 61.389Q-82.103 61.386-82.096 61.384Q-82.089 61.382-82.079 61.382L-82.028 61.382Q-82.001 61.382-81.977 61.406Q-81.953 61.430-81.953 61.458L-81.953 62.305Q-81.953 62.326-81.977 62.353Q-82.001 62.380-82.028 62.380L-82.141 62.380Q-82.168 62.380-82.194 62.355Q-82.219 62.329-82.219 62.305Q-82.219 62.069-82.325 61.905Q-82.431 61.741-82.614 61.659Q-82.797 61.577-83.029 61.577Q-83.357 61.577-83.614 61.680Q-83.870 61.782-83.870 62.059Q-83.870 62.254-83.687 62.363Q-83.504 62.473-83.275 62.514L-82.701 62.620Q-82.455 62.668-82.242 62.796Q-82.028 62.924-81.891 63.127Q-81.754 63.331-81.754 63.580Q-81.754 64.093-82.120 64.332Q-82.486 64.571-83.023 64.571Q-83.518 64.571-83.850 64.277L-84.116 64.551Q-84.137 64.571-84.164 64.571L-84.212 64.571Q-84.236 64.571-84.263 64.544Q-84.291 64.517-84.291 64.496M-81.126 64.496L-81.126 63.433Q-81.126 63.409-81.098 63.382Q-81.071 63.355-81.047 63.355L-80.938 63.355Q-80.873 63.355-80.859 63.413Q-80.763 63.847-80.517 64.098Q-80.271 64.349-79.857 64.349Q-79.516 64.349-79.263 64.216Q-79.010 64.083-79.010 63.775Q-79.010 63.618-79.104 63.503Q-79.198 63.389-79.336 63.320Q-79.475 63.252-79.642 63.214L-80.223 63.115Q-80.579 63.047-80.852 62.826Q-81.126 62.606-81.126 62.264Q-81.126 62.015-81.014 61.840Q-80.903 61.666-80.717 61.567Q-80.531 61.468-80.316 61.425Q-80.100 61.382-79.857 61.382Q-79.444 61.382-79.164 61.564L-78.948 61.389Q-78.938 61.386-78.931 61.384Q-78.924 61.382-78.914 61.382L-78.863 61.382Q-78.836 61.382-78.812 61.406Q-78.788 61.430-78.788 61.458L-78.788 62.305Q-78.788 62.326-78.812 62.353Q-78.836 62.380-78.863 62.380L-78.976 62.380Q-79.003 62.380-79.029 62.355Q-79.054 62.329-79.054 62.305Q-79.054 62.069-79.160 61.905Q-79.266 61.741-79.449 61.659Q-79.632 61.577-79.864 61.577Q-80.192 61.577-80.449 61.680Q-80.705 61.782-80.705 62.059Q-80.705 62.254-80.522 62.363Q-80.339 62.473-80.110 62.514L-79.536 62.620Q-79.290 62.668-79.076 62.796Q-78.863 62.924-78.726 63.127Q-78.589 63.331-78.589 63.580Q-78.589 64.093-78.955 64.332Q-79.321 64.571-79.857 64.571Q-80.353 64.571-80.685 64.277L-80.951 64.551Q-80.972 64.571-80.999 64.571L-81.047 64.571Q-81.071 64.571-81.098 64.544Q-81.126 64.517-81.126 64.496\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(218.35 -83.608)\">\u003Cpath d=\"M-121.188 64.503L-121.455 64.503L-121.455 60.395Q-121.455 60.125-121.562 60.063Q-121.670 60.002-121.981 60.002L-121.981 59.721L-120.901 59.646L-120.901 61.816Q-120.692 61.625-120.407 61.521Q-120.121 61.417-119.824 61.417Q-119.506 61.417-119.209 61.538Q-118.912 61.659-118.689 61.875Q-118.467 62.090-118.341 62.375Q-118.214 62.661-118.214 62.992Q-118.214 63.437-118.454 63.801Q-118.693 64.165-119.086 64.368Q-119.479 64.571-119.923 64.571Q-120.118 64.571-120.308 64.515Q-120.497 64.459-120.658 64.354Q-120.819 64.250-120.959 64.089L-121.188 64.503M-120.873 62.158L-120.873 63.775Q-120.737 64.035-120.496 64.192Q-120.255 64.349-119.978 64.349Q-119.684 64.349-119.472 64.242Q-119.260 64.134-119.127 63.942Q-118.994 63.751-118.935 63.512Q-118.877 63.273-118.877 62.992Q-118.877 62.633-118.971 62.329Q-119.065 62.025-119.293 61.832Q-119.520 61.639-119.886 61.639Q-120.186 61.639-120.453 61.775Q-120.720 61.912-120.873 62.158M-117.520 63.775Q-117.520 63.443-117.297 63.216Q-117.073 62.989-116.729 62.861Q-116.386 62.732-116.013 62.680Q-115.641 62.627-115.336 62.627L-115.336 62.374Q-115.336 62.169-115.444 61.989Q-115.552 61.810-115.733 61.707Q-115.914 61.605-116.122 61.605Q-116.529 61.605-116.765 61.697Q-116.676 61.734-116.630 61.818Q-116.584 61.902-116.584 62.004Q-116.584 62.100-116.630 62.179Q-116.676 62.257-116.757 62.302Q-116.837 62.346-116.926 62.346Q-117.076 62.346-117.177 62.249Q-117.278 62.151-117.278 62.004Q-117.278 61.382-116.122 61.382Q-115.911 61.382-115.661 61.446Q-115.412 61.509-115.210 61.628Q-115.008 61.748-114.882 61.933Q-114.755 62.117-114.755 62.360L-114.755 63.936Q-114.755 64.052-114.694 64.148Q-114.632 64.243-114.519 64.243Q-114.410 64.243-114.345 64.149Q-114.280 64.055-114.280 63.936L-114.280 63.488L-114.014 63.488L-114.014 63.936Q-114.014 64.206-114.241 64.371Q-114.468 64.537-114.748 64.537Q-114.957 64.537-115.094 64.383Q-115.230 64.230-115.254 64.014Q-115.401 64.281-115.683 64.426Q-115.965 64.571-116.290 64.571Q-116.567 64.571-116.851 64.496Q-117.134 64.421-117.327 64.242Q-117.520 64.062-117.520 63.775M-116.905 63.775Q-116.905 63.949-116.804 64.079Q-116.704 64.209-116.548 64.279Q-116.392 64.349-116.228 64.349Q-116.010 64.349-115.801 64.252Q-115.593 64.154-115.465 63.973Q-115.336 63.792-115.336 63.566L-115.336 62.838Q-115.661 62.838-116.027 62.929Q-116.392 63.020-116.649 63.232Q-116.905 63.443-116.905 63.775M-113.597 62.992Q-113.597 62.664-113.462 62.363Q-113.327 62.063-113.091 61.842Q-112.855 61.622-112.551 61.502Q-112.246 61.382-111.922 61.382Q-111.416 61.382-111.067 61.485Q-110.719 61.587-110.719 61.963Q-110.719 62.110-110.816 62.211Q-110.913 62.312-111.060 62.312Q-111.214 62.312-111.313 62.213Q-111.413 62.114-111.413 61.963Q-111.413 61.775-111.272 61.683Q-111.474 61.632-111.915 61.632Q-112.270 61.632-112.499 61.828Q-112.728 62.025-112.829 62.334Q-112.930 62.644-112.930 62.992Q-112.930 63.341-112.804 63.647Q-112.677 63.953-112.423 64.137Q-112.168 64.322-111.812 64.322Q-111.590 64.322-111.406 64.238Q-111.221 64.154-111.086 63.999Q-110.951 63.843-110.893 63.635Q-110.879 63.580-110.825 63.580L-110.712 63.580Q-110.681 63.580-110.659 63.604Q-110.637 63.628-110.637 63.662L-110.637 63.683Q-110.722 63.970-110.910 64.168Q-111.098 64.366-111.363 64.469Q-111.628 64.571-111.922 64.571Q-112.352 64.571-112.740 64.365Q-113.128 64.158-113.362 63.795Q-113.597 63.433-113.597 62.992\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(218.35 -83.608)\">\u003Cpath d=\"M-108.655 64.503L-110.238 64.503L-110.238 64.223Q-110.009 64.223-109.860 64.189Q-109.712 64.154-109.712 64.014L-109.712 60.395Q-109.712 60.125-109.819 60.063Q-109.927 60.002-110.238 60.002L-110.238 59.721L-109.158 59.646L-109.158 62.934L-108.173 62.165Q-107.968 62.028-107.968 61.878Q-107.968 61.834-108.009 61.799Q-108.050 61.765-108.095 61.765L-108.095 61.485L-106.731 61.485L-106.731 61.765Q-107.220 61.765-107.739 62.165L-108.296 62.599L-107.319 63.823Q-107.117 64.069-106.984 64.146Q-106.851 64.223-106.564 64.223L-106.564 64.503L-107.996 64.503L-107.996 64.223Q-107.808 64.223-107.808 64.110Q-107.808 64.014-107.962 63.823L-108.696 62.914L-109.178 63.293L-109.178 64.014Q-109.178 64.151-109.030 64.187Q-108.881 64.223-108.655 64.223L-108.655 64.503M-104.663 64.476L-105.644 61.977Q-105.706 61.834-105.824 61.799Q-105.941 61.765-106.157 61.765L-106.157 61.485L-104.677 61.485L-104.677 61.765Q-105.056 61.765-105.056 61.926Q-105.056 61.936-105.043 61.977L-104.328 63.809L-103.655 62.104Q-103.686 62.032-103.686 62.004Q-103.686 61.977-103.713 61.977Q-103.774 61.830-103.892 61.798Q-104.010 61.765-104.222 61.765L-104.222 61.485L-102.824 61.485L-102.824 61.765Q-103.200 61.765-103.200 61.926Q-103.200 61.957-103.193 61.977L-102.438 63.915L-101.751 62.165Q-101.731 62.114-101.731 62.059Q-101.731 61.919-101.843 61.842Q-101.956 61.765-102.096 61.765L-102.096 61.485L-100.876 61.485L-100.876 61.765Q-101.081 61.765-101.237 61.871Q-101.392 61.977-101.464 62.165L-102.370 64.476Q-102.404 64.571-102.517 64.571L-102.585 64.571Q-102.694 64.571-102.732 64.476L-103.515 62.473L-104.301 64.476Q-104.335 64.571-104.448 64.571L-104.516 64.571Q-104.626 64.571-104.663 64.476\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(218.35 -83.608)\">\u003Cpath d=\"M-100.498 63.775Q-100.498 63.443-100.275 63.216Q-100.051 62.989-99.707 62.861Q-99.364 62.732-98.991 62.680Q-98.619 62.627-98.314 62.627L-98.314 62.374Q-98.314 62.169-98.422 61.989Q-98.530 61.810-98.711 61.707Q-98.892 61.605-99.100 61.605Q-99.507 61.605-99.743 61.697Q-99.654 61.734-99.608 61.818Q-99.562 61.902-99.562 62.004Q-99.562 62.100-99.608 62.179Q-99.654 62.257-99.735 62.302Q-99.815 62.346-99.904 62.346Q-100.054 62.346-100.155 62.249Q-100.256 62.151-100.256 62.004Q-100.256 61.382-99.100 61.382Q-98.889 61.382-98.639 61.446Q-98.390 61.509-98.188 61.628Q-97.986 61.748-97.860 61.933Q-97.733 62.117-97.733 62.360L-97.733 63.936Q-97.733 64.052-97.672 64.148Q-97.610 64.243-97.497 64.243Q-97.388 64.243-97.323 64.149Q-97.258 64.055-97.258 63.936L-97.258 63.488L-96.992 63.488L-96.992 63.936Q-96.992 64.206-97.219 64.371Q-97.446 64.537-97.726 64.537Q-97.935 64.537-98.072 64.383Q-98.208 64.230-98.232 64.014Q-98.379 64.281-98.661 64.426Q-98.943 64.571-99.268 64.571Q-99.545 64.571-99.829 64.496Q-100.112 64.421-100.305 64.242Q-100.498 64.062-100.498 63.775M-99.883 63.775Q-99.883 63.949-99.782 64.079Q-99.682 64.209-99.526 64.279Q-99.371 64.349-99.206 64.349Q-98.988 64.349-98.779 64.252Q-98.571 64.154-98.443 63.973Q-98.314 63.792-98.314 63.566L-98.314 62.838Q-98.639 62.838-99.005 62.929Q-99.371 63.020-99.627 63.232Q-99.883 63.443-99.883 63.775M-94.825 64.503L-96.561 64.503L-96.561 64.223Q-96.332 64.223-96.183 64.189Q-96.035 64.154-96.035 64.014L-96.035 62.165Q-96.035 61.895-96.142 61.834Q-96.250 61.772-96.561 61.772L-96.561 61.492L-95.532 61.417L-95.532 62.124Q-95.402 61.816-95.160 61.617Q-94.917 61.417-94.599 61.417Q-94.380 61.417-94.209 61.541Q-94.038 61.666-94.038 61.878Q-94.038 62.015-94.138 62.114Q-94.237 62.213-94.370 62.213Q-94.507 62.213-94.606 62.114Q-94.705 62.015-94.705 61.878Q-94.705 61.738-94.606 61.639Q-94.896 61.639-95.096 61.835Q-95.296 62.032-95.389 62.326Q-95.481 62.620-95.481 62.900L-95.481 64.014Q-95.481 64.223-94.825 64.223L-94.825 64.503M-93.454 62.992Q-93.454 62.654-93.314 62.363Q-93.174 62.073-92.929 61.859Q-92.685 61.646-92.381 61.531Q-92.077 61.417-91.752 61.417Q-91.482 61.417-91.219 61.516Q-90.955 61.615-90.764 61.793L-90.764 60.395Q-90.764 60.125-90.872 60.063Q-90.979 60.002-91.290 60.002L-91.290 59.721L-90.214 59.646L-90.214 63.830Q-90.214 64.018-90.159 64.101Q-90.104 64.185-90.004 64.204Q-89.903 64.223-89.687 64.223L-89.687 64.503L-90.795 64.571L-90.795 64.154Q-91.212 64.571-91.837 64.571Q-92.268 64.571-92.641 64.359Q-93.013 64.148-93.234 63.787Q-93.454 63.426-93.454 62.992M-91.779 64.349Q-91.571 64.349-91.384 64.277Q-91.198 64.206-91.044 64.069Q-90.891 63.932-90.795 63.754L-90.795 62.145Q-90.880 61.998-91.026 61.878Q-91.171 61.758-91.340 61.699Q-91.509 61.639-91.690 61.639Q-92.251 61.639-92.519 62.028Q-92.788 62.418-92.788 62.999Q-92.788 63.570-92.553 63.960Q-92.319 64.349-91.779 64.349\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(218.35 -83.608)\">\u003Cpath d=\"M-84.691 65.860L-86.321 65.860L-86.321 65.580Q-86.092 65.580-85.943 65.545Q-85.795 65.511-85.795 65.371L-85.795 62.025Q-85.795 61.854-85.931 61.813Q-86.068 61.772-86.321 61.772L-86.321 61.492L-85.241 61.417L-85.241 61.823Q-85.019 61.622-84.732 61.519Q-84.444 61.417-84.137 61.417Q-83.710 61.417-83.346 61.630Q-82.982 61.844-82.768 62.208Q-82.554 62.572-82.554 62.992Q-82.554 63.437-82.794 63.801Q-83.033 64.165-83.426 64.368Q-83.819 64.571-84.263 64.571Q-84.530 64.571-84.778 64.471Q-85.025 64.370-85.213 64.189L-85.213 65.371Q-85.213 65.508-85.065 65.544Q-84.916 65.580-84.691 65.580L-84.691 65.860M-85.213 62.172L-85.213 63.782Q-85.080 64.035-84.837 64.192Q-84.595 64.349-84.318 64.349Q-83.990 64.349-83.737 64.148Q-83.484 63.946-83.351 63.628Q-83.217 63.310-83.217 62.992Q-83.217 62.763-83.282 62.534Q-83.347 62.305-83.475 62.107Q-83.604 61.909-83.798 61.789Q-83.993 61.670-84.226 61.670Q-84.520 61.670-84.788 61.799Q-85.056 61.929-85.213 62.172M-81.860 63.775Q-81.860 63.443-81.637 63.216Q-81.413 62.989-81.069 62.861Q-80.726 62.732-80.353 62.680Q-79.981 62.627-79.676 62.627L-79.676 62.374Q-79.676 62.169-79.784 61.989Q-79.892 61.810-80.073 61.707Q-80.254 61.605-80.462 61.605Q-80.869 61.605-81.105 61.697Q-81.016 61.734-80.970 61.818Q-80.924 61.902-80.924 62.004Q-80.924 62.100-80.970 62.179Q-81.016 62.257-81.097 62.302Q-81.177 62.346-81.266 62.346Q-81.416 62.346-81.517 62.249Q-81.618 62.151-81.618 62.004Q-81.618 61.382-80.462 61.382Q-80.251 61.382-80.001 61.446Q-79.752 61.509-79.550 61.628Q-79.348 61.748-79.222 61.933Q-79.095 62.117-79.095 62.360L-79.095 63.936Q-79.095 64.052-79.034 64.148Q-78.972 64.243-78.859 64.243Q-78.750 64.243-78.685 64.149Q-78.620 64.055-78.620 63.936L-78.620 63.488L-78.354 63.488L-78.354 63.936Q-78.354 64.206-78.581 64.371Q-78.808 64.537-79.088 64.537Q-79.297 64.537-79.434 64.383Q-79.570 64.230-79.594 64.014Q-79.741 64.281-80.023 64.426Q-80.305 64.571-80.630 64.571Q-80.907 64.571-81.191 64.496Q-81.474 64.421-81.667 64.242Q-81.860 64.062-81.860 63.775M-81.245 63.775Q-81.245 63.949-81.144 64.079Q-81.044 64.209-80.888 64.279Q-80.732 64.349-80.568 64.349Q-80.350 64.349-80.141 64.252Q-79.933 64.154-79.805 63.973Q-79.676 63.792-79.676 63.566L-79.676 62.838Q-80.001 62.838-80.367 62.929Q-80.732 63.020-80.989 63.232Q-81.245 63.443-81.245 63.775M-77.937 64.496L-77.937 63.433Q-77.937 63.409-77.909 63.382Q-77.882 63.355-77.858 63.355L-77.749 63.355Q-77.684 63.355-77.670 63.413Q-77.574 63.847-77.328 64.098Q-77.082 64.349-76.669 64.349Q-76.327 64.349-76.074 64.216Q-75.821 64.083-75.821 63.775Q-75.821 63.618-75.915 63.503Q-76.009 63.389-76.147 63.320Q-76.286 63.252-76.453 63.214L-77.034 63.115Q-77.390 63.047-77.663 62.826Q-77.937 62.606-77.937 62.264Q-77.937 62.015-77.826 61.840Q-77.714 61.666-77.528 61.567Q-77.342 61.468-77.127 61.425Q-76.911 61.382-76.669 61.382Q-76.255 61.382-75.975 61.564L-75.759 61.389Q-75.749 61.386-75.742 61.384Q-75.735 61.382-75.725 61.382L-75.674 61.382Q-75.647 61.382-75.623 61.406Q-75.599 61.430-75.599 61.458L-75.599 62.305Q-75.599 62.326-75.623 62.353Q-75.647 62.380-75.674 62.380L-75.787 62.380Q-75.814 62.380-75.840 62.355Q-75.865 62.329-75.865 62.305Q-75.865 62.069-75.971 61.905Q-76.077 61.741-76.260 61.659Q-76.443 61.577-76.675 61.577Q-77.003 61.577-77.260 61.680Q-77.516 61.782-77.516 62.059Q-77.516 62.254-77.333 62.363Q-77.150 62.473-76.921 62.514L-76.347 62.620Q-76.101 62.668-75.888 62.796Q-75.674 62.924-75.537 63.127Q-75.400 63.331-75.400 63.580Q-75.400 64.093-75.766 64.332Q-76.132 64.571-76.669 64.571Q-77.164 64.571-77.496 64.277L-77.762 64.551Q-77.783 64.571-77.810 64.571L-77.858 64.571Q-77.882 64.571-77.909 64.544Q-77.937 64.517-77.937 64.496M-74.772 64.496L-74.772 63.433Q-74.772 63.409-74.744 63.382Q-74.717 63.355-74.693 63.355L-74.584 63.355Q-74.519 63.355-74.505 63.413Q-74.409 63.847-74.163 64.098Q-73.917 64.349-73.503 64.349Q-73.162 64.349-72.909 64.216Q-72.656 64.083-72.656 63.775Q-72.656 63.618-72.750 63.503Q-72.844 63.389-72.982 63.320Q-73.121 63.252-73.288 63.214L-73.869 63.115Q-74.225 63.047-74.498 62.826Q-74.772 62.606-74.772 62.264Q-74.772 62.015-74.660 61.840Q-74.549 61.666-74.363 61.567Q-74.177 61.468-73.962 61.425Q-73.746 61.382-73.503 61.382Q-73.090 61.382-72.810 61.564L-72.594 61.389Q-72.584 61.386-72.577 61.384Q-72.570 61.382-72.560 61.382L-72.509 61.382Q-72.482 61.382-72.458 61.406Q-72.434 61.430-72.434 61.458L-72.434 62.305Q-72.434 62.326-72.458 62.353Q-72.482 62.380-72.509 62.380L-72.622 62.380Q-72.649 62.380-72.675 62.355Q-72.700 62.329-72.700 62.305Q-72.700 62.069-72.806 61.905Q-72.912 61.741-73.095 61.659Q-73.278 61.577-73.510 61.577Q-73.838 61.577-74.095 61.680Q-74.351 61.782-74.351 62.059Q-74.351 62.254-74.168 62.363Q-73.985 62.473-73.756 62.514L-73.182 62.620Q-72.936 62.668-72.722 62.796Q-72.509 62.924-72.372 63.127Q-72.235 63.331-72.235 63.580Q-72.235 64.093-72.601 64.332Q-72.967 64.571-73.503 64.571Q-73.999 64.571-74.331 64.277L-74.597 64.551Q-74.618 64.571-74.645 64.571L-74.693 64.571Q-74.717 64.571-74.744 64.544Q-74.772 64.517-74.772 64.496\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fsvg>\u003Cfigcaption class=\"tikz-cap\">A bidirectional RNN. A forward RNN and a backward RNN process the sequence in opposite directions; their hidden states are concatenated at each position so every output sees both the left and the right context.\u003C\u002Ffigcaption>","\u003Csvg style=\"width:100%;max-width:518.190px;height:auto\" xmlns=\"http:\u002F\u002Fwww.w3.org\u002F2000\u002Fsvg\" viewBox=\"-75 -75 388.643 138.145\">\u003Cg stroke=\"currentColor\" style=\"stroke-miterlimit:10;stroke-width:.4\">\u003Cpath fill=\"none\" d=\"M-65.403 33.205h28.452V13.288h-28.452Z\"\u002F>\u003Cg transform=\"translate(53.996 2.43)\">\u003Cpath d=\"M-106.100 23.247L-107.703 23.247L-107.703 22.967Q-107.477 22.967-107.328 22.933Q-107.180 22.898-107.180 22.758L-107.180 19.139Q-107.180 18.869-107.287 18.807Q-107.395 18.746-107.703 18.746L-107.703 18.465L-106.626 18.390L-106.626 22.758Q-106.626 22.895-106.476 22.931Q-106.325 22.967-106.100 22.967L-106.100 23.247M-105.546 21.712Q-105.546 21.391-105.421 21.102Q-105.296 20.813-105.071 20.590Q-104.845 20.366-104.550 20.246Q-104.254 20.126-103.936 20.126Q-103.608 20.126-103.346 20.226Q-103.085 20.325-102.909 20.507Q-102.733 20.690-102.639 20.948Q-102.545 21.206-102.545 21.538Q-102.545 21.630-102.627 21.651L-104.883 21.651L-104.883 21.712Q-104.883 22.300-104.599 22.683Q-104.315 23.066-103.748 23.066Q-103.427 23.066-103.158 22.873Q-102.890 22.680-102.801 22.365Q-102.794 22.324-102.719 22.310L-102.627 22.310Q-102.545 22.334-102.545 22.406Q-102.545 22.413-102.552 22.440Q-102.665 22.837-103.035 23.076Q-103.406 23.315-103.830 23.315Q-104.268 23.315-104.667 23.107Q-105.067 22.898-105.307 22.531Q-105.546 22.164-105.546 21.712M-104.876 21.442L-103.061 21.442Q-103.061 21.165-103.158 20.913Q-103.256 20.660-103.454 20.504Q-103.652 20.349-103.936 20.349Q-104.213 20.349-104.426 20.507Q-104.640 20.666-104.758 20.921Q-104.876 21.176-104.876 21.442\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-65.403-9.474h28.452v-19.917h-28.452Z\"\u002F>\u003Cg transform=\"translate(51.142 -41.172)\">\u003Cpath d=\"M-107.809 21.712Q-107.809 21.391-107.684 21.102Q-107.559 20.813-107.333 20.590Q-107.108 20.366-106.812 20.246Q-106.517 20.126-106.199 20.126Q-105.871 20.126-105.609 20.226Q-105.348 20.325-105.172 20.507Q-104.996 20.690-104.902 20.948Q-104.808 21.206-104.808 21.538Q-104.808 21.630-104.890 21.651L-107.145 21.651L-107.145 21.712Q-107.145 22.300-106.862 22.683Q-106.578 23.066-106.011 23.066Q-105.689 23.066-105.421 22.873Q-105.153 22.680-105.064 22.365Q-105.057 22.324-104.982 22.310L-104.890 22.310Q-104.808 22.334-104.808 22.406Q-104.808 22.413-104.814 22.440Q-104.927 22.837-105.298 23.076Q-105.669 23.315-106.093 23.315Q-106.530 23.315-106.930 23.107Q-107.330 22.898-107.569 22.531Q-107.809 22.164-107.809 21.712M-107.139 21.442L-105.324 21.442Q-105.324 21.165-105.421 20.913Q-105.519 20.660-105.717 20.504Q-105.915 20.349-106.199 20.349Q-106.476 20.349-106.689 20.507Q-106.903 20.666-107.021 20.921Q-107.139 21.176-107.139 21.442M-102.538 23.247L-104.172 23.247L-104.172 22.967Q-103.943 22.967-103.794 22.933Q-103.645 22.898-103.645 22.758L-103.645 20.909Q-103.645 20.639-103.753 20.578Q-103.861 20.516-104.172 20.516L-104.172 20.236L-103.112 20.161L-103.112 20.810Q-102.941 20.502-102.637 20.331Q-102.333 20.161-101.988 20.161Q-101.482 20.161-101.198 20.384Q-100.915 20.608-100.915 21.104L-100.915 22.758Q-100.915 22.895-100.766 22.931Q-100.617 22.967-100.392 22.967L-100.392 23.247L-102.022 23.247L-102.022 22.967Q-101.793 22.967-101.644 22.933Q-101.496 22.898-101.496 22.758L-101.496 21.118Q-101.496 20.783-101.615 20.583Q-101.735 20.383-102.049 20.383Q-102.319 20.383-102.553 20.519Q-102.788 20.656-102.926 20.890Q-103.064 21.124-103.064 21.398L-103.064 22.758Q-103.064 22.895-102.914 22.931Q-102.764 22.967-102.538 22.967L-102.538 23.247M-99.804 21.736Q-99.804 21.408-99.669 21.107Q-99.534 20.807-99.298 20.586Q-99.062 20.366-98.758 20.246Q-98.454 20.126-98.129 20.126Q-97.623 20.126-97.274 20.229Q-96.926 20.331-96.926 20.707Q-96.926 20.854-97.023 20.955Q-97.121 21.056-97.268 21.056Q-97.421 21.056-97.520 20.957Q-97.620 20.858-97.620 20.707Q-97.620 20.519-97.479 20.427Q-97.681 20.376-98.122 20.376Q-98.478 20.376-98.707 20.572Q-98.936 20.769-99.036 21.078Q-99.137 21.388-99.137 21.736Q-99.137 22.085-99.011 22.391Q-98.884 22.697-98.630 22.881Q-98.375 23.066-98.019 23.066Q-97.797 23.066-97.613 22.982Q-97.428 22.898-97.293 22.743Q-97.158 22.587-97.100 22.379Q-97.086 22.324-97.032 22.324L-96.919 22.324Q-96.888 22.324-96.866 22.348Q-96.844 22.372-96.844 22.406L-96.844 22.427Q-96.929 22.714-97.117 22.912Q-97.305 23.110-97.570 23.213Q-97.835 23.315-98.129 23.315Q-98.560 23.315-98.947 23.109Q-99.335 22.902-99.570 22.539Q-99.804 22.177-99.804 21.736\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-51.177 13.088V-6.674\"\u002F>\u003Cpath stroke=\"none\" d=\"m-51.177-9.274-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-8.498 33.205h28.453V13.288H-8.498Z\"\u002F>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(106.377 2.43)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.408-107.633 21.107Q-107.498 20.807-107.262 20.586Q-107.026 20.366-106.722 20.246Q-106.417 20.126-106.093 20.126Q-105.587 20.126-105.238 20.229Q-104.890 20.331-104.890 20.707Q-104.890 20.854-104.987 20.955Q-105.084 21.056-105.231 21.056Q-105.385 21.056-105.484 20.957Q-105.583 20.858-105.583 20.707Q-105.583 20.519-105.443 20.427Q-105.645 20.376-106.086 20.376Q-106.441 20.376-106.670 20.572Q-106.899 20.769-107 21.078Q-107.101 21.388-107.101 21.736Q-107.101 22.085-106.975 22.391Q-106.848 22.697-106.593 22.881Q-106.339 23.066-105.983 23.066Q-105.761 23.066-105.577 22.982Q-105.392 22.898-105.257 22.743Q-105.122 22.587-105.064 22.379Q-105.050 22.324-104.996 22.324L-104.883 22.324Q-104.852 22.324-104.830 22.348Q-104.808 22.372-104.808 22.406L-104.808 22.427Q-104.893 22.714-105.081 22.912Q-105.269 23.110-105.534 23.213Q-105.799 23.315-106.093 23.315Q-106.523 23.315-106.911 23.109Q-107.299 22.902-107.533 22.539Q-107.768 22.177-107.768 21.736\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(106.377 2.43)\">\u003Cpath d=\"M-102.746 23.247L-104.380 23.247L-104.380 22.967Q-104.151 22.967-104.002 22.933Q-103.853 22.898-103.853 22.758L-103.853 19.139Q-103.853 18.869-103.961 18.807Q-104.069 18.746-104.380 18.746L-104.380 18.465L-103.300 18.390L-103.300 20.776Q-103.194 20.591-103.016 20.449Q-102.838 20.308-102.630 20.234Q-102.421 20.161-102.196 20.161Q-101.690 20.161-101.406 20.384Q-101.122 20.608-101.122 21.104L-101.122 22.758Q-101.122 22.895-100.974 22.931Q-100.825 22.967-100.599 22.967L-100.599 23.247L-102.230 23.247L-102.230 22.967Q-102.001 22.967-101.852 22.933Q-101.703 22.898-101.703 22.758L-101.703 21.118Q-101.703 20.783-101.823 20.583Q-101.943 20.383-102.257 20.383Q-102.527 20.383-102.761 20.519Q-102.995 20.656-103.134 20.890Q-103.272 21.124-103.272 21.398L-103.272 22.758Q-103.272 22.895-103.122 22.931Q-102.971 22.967-102.746 22.967L-102.746 23.247M-99.953 22.519Q-99.953 22.187-99.730 21.960Q-99.506 21.733-99.162 21.605Q-98.819 21.476-98.446 21.424Q-98.074 21.371-97.769 21.371L-97.769 21.118Q-97.769 20.913-97.877 20.733Q-97.985 20.554-98.166 20.451Q-98.347 20.349-98.555 20.349Q-98.962 20.349-99.198 20.441Q-99.109 20.478-99.063 20.562Q-99.017 20.646-99.017 20.748Q-99.017 20.844-99.063 20.923Q-99.109 21.001-99.190 21.046Q-99.270 21.090-99.359 21.090Q-99.509 21.090-99.610 20.993Q-99.711 20.895-99.711 20.748Q-99.711 20.126-98.555 20.126Q-98.344 20.126-98.094 20.190Q-97.845 20.253-97.643 20.372Q-97.441 20.492-97.315 20.677Q-97.188 20.861-97.188 21.104L-97.188 22.680Q-97.188 22.796-97.127 22.892Q-97.065 22.987-96.952 22.987Q-96.843 22.987-96.778 22.893Q-96.713 22.799-96.713 22.680L-96.713 22.232L-96.447 22.232L-96.447 22.680Q-96.447 22.950-96.674 23.115Q-96.901 23.281-97.181 23.281Q-97.390 23.281-97.527 23.127Q-97.663 22.974-97.687 22.758Q-97.834 23.025-98.116 23.170Q-98.398 23.315-98.723 23.315Q-99 23.315-99.284 23.240Q-99.567 23.165-99.760 22.986Q-99.953 22.806-99.953 22.519M-99.338 22.519Q-99.338 22.693-99.237 22.823Q-99.137 22.953-98.981 23.023Q-98.825 23.093-98.661 23.093Q-98.443 23.093-98.234 22.996Q-98.026 22.898-97.898 22.717Q-97.769 22.536-97.769 22.310L-97.769 21.582Q-98.094 21.582-98.460 21.673Q-98.825 21.764-99.082 21.976Q-99.338 22.187-99.338 22.519M-95.503 22.406L-95.503 20.509L-96.142 20.509L-96.142 20.287Q-95.825 20.287-95.607 20.077Q-95.390 19.867-95.290 19.557Q-95.189 19.248-95.189 18.940L-94.922 18.940L-94.922 20.229L-93.846 20.229L-93.846 20.509L-94.922 20.509L-94.922 22.393Q-94.922 22.669-94.818 22.868Q-94.714 23.066-94.454 23.066Q-94.297 23.066-94.191 22.962Q-94.085 22.857-94.035 22.704Q-93.986 22.550-93.986 22.393L-93.986 21.979L-93.719 21.979L-93.719 22.406Q-93.719 22.632-93.818 22.842Q-93.917 23.052-94.102 23.184Q-94.286 23.315-94.515 23.315Q-94.953 23.315-95.228 23.078Q-95.503 22.840-95.503 22.406\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M-8.498-9.474h28.453v-19.917H-8.498Z\"\u002F>\u003Cg transform=\"translate(108.047 -41.172)\">\u003Cpath d=\"M-107.809 21.712Q-107.809 21.391-107.684 21.102Q-107.559 20.813-107.333 20.590Q-107.108 20.366-106.812 20.246Q-106.517 20.126-106.199 20.126Q-105.871 20.126-105.609 20.226Q-105.348 20.325-105.172 20.507Q-104.996 20.690-104.902 20.948Q-104.808 21.206-104.808 21.538Q-104.808 21.630-104.890 21.651L-107.145 21.651L-107.145 21.712Q-107.145 22.300-106.862 22.683Q-106.578 23.066-106.011 23.066Q-105.689 23.066-105.421 22.873Q-105.153 22.680-105.064 22.365Q-105.057 22.324-104.982 22.310L-104.890 22.310Q-104.808 22.334-104.808 22.406Q-104.808 22.413-104.814 22.440Q-104.927 22.837-105.298 23.076Q-105.669 23.315-106.093 23.315Q-106.530 23.315-106.930 23.107Q-107.330 22.898-107.569 22.531Q-107.809 22.164-107.809 21.712M-107.139 21.442L-105.324 21.442Q-105.324 21.165-105.421 20.913Q-105.519 20.660-105.717 20.504Q-105.915 20.349-106.199 20.349Q-106.476 20.349-106.689 20.507Q-106.903 20.666-107.021 20.921Q-107.139 21.176-107.139 21.442M-102.538 23.247L-104.172 23.247L-104.172 22.967Q-103.943 22.967-103.794 22.933Q-103.645 22.898-103.645 22.758L-103.645 20.909Q-103.645 20.639-103.753 20.578Q-103.861 20.516-104.172 20.516L-104.172 20.236L-103.112 20.161L-103.112 20.810Q-102.941 20.502-102.637 20.331Q-102.333 20.161-101.988 20.161Q-101.482 20.161-101.198 20.384Q-100.915 20.608-100.915 21.104L-100.915 22.758Q-100.915 22.895-100.766 22.931Q-100.617 22.967-100.392 22.967L-100.392 23.247L-102.022 23.247L-102.022 22.967Q-101.793 22.967-101.644 22.933Q-101.496 22.898-101.496 22.758L-101.496 21.118Q-101.496 20.783-101.615 20.583Q-101.735 20.383-102.049 20.383Q-102.319 20.383-102.553 20.519Q-102.788 20.656-102.926 20.890Q-103.064 21.124-103.064 21.398L-103.064 22.758Q-103.064 22.895-102.914 22.931Q-102.764 22.967-102.538 22.967L-102.538 23.247M-99.804 21.736Q-99.804 21.408-99.669 21.107Q-99.534 20.807-99.298 20.586Q-99.062 20.366-98.758 20.246Q-98.454 20.126-98.129 20.126Q-97.623 20.126-97.274 20.229Q-96.926 20.331-96.926 20.707Q-96.926 20.854-97.023 20.955Q-97.121 21.056-97.268 21.056Q-97.421 21.056-97.520 20.957Q-97.620 20.858-97.620 20.707Q-97.620 20.519-97.479 20.427Q-97.681 20.376-98.122 20.376Q-98.478 20.376-98.707 20.572Q-98.936 20.769-99.036 21.078Q-99.137 21.388-99.137 21.736Q-99.137 22.085-99.011 22.391Q-98.884 22.697-98.630 22.881Q-98.375 23.066-98.019 23.066Q-97.797 23.066-97.613 22.982Q-97.428 22.898-97.293 22.743Q-97.158 22.587-97.100 22.379Q-97.086 22.324-97.032 22.324L-96.919 22.324Q-96.888 22.324-96.866 22.348Q-96.844 22.372-96.844 22.406L-96.844 22.427Q-96.929 22.714-97.117 22.912Q-97.305 23.110-97.570 23.213Q-97.835 23.315-98.129 23.315Q-98.560 23.315-98.947 23.109Q-99.335 22.902-99.570 22.539Q-99.804 22.177-99.804 21.736\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M5.728 13.088V-6.674\"\u002F>\u003Cpath stroke=\"none\" d=\"m5.728-9.274-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M48.408 33.205H76.86V13.288H48.408Z\"\u002F>\u003Cg transform=\"translate(163.39 2.43)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.398-107.627 21.107Q-107.487 20.817-107.243 20.603Q-106.999 20.390-106.694 20.275Q-106.390 20.161-106.065 20.161Q-105.795 20.161-105.532 20.260Q-105.269 20.359-105.078 20.537L-105.078 19.139Q-105.078 18.869-105.185 18.807Q-105.293 18.746-105.604 18.746L-105.604 18.465L-104.527 18.390L-104.527 22.574Q-104.527 22.762-104.473 22.845Q-104.418 22.929-104.317 22.948Q-104.216 22.967-104.001 22.967L-104.001 23.247L-105.108 23.315L-105.108 22.898Q-105.525 23.315-106.151 23.315Q-106.582 23.315-106.954 23.103Q-107.327 22.892-107.547 22.531Q-107.768 22.170-107.768 21.736M-106.093 23.093Q-105.884 23.093-105.698 23.021Q-105.512 22.950-105.358 22.813Q-105.204 22.676-105.108 22.498L-105.108 20.889Q-105.194 20.742-105.339 20.622Q-105.484 20.502-105.654 20.443Q-105.823 20.383-106.004 20.383Q-106.564 20.383-106.833 20.772Q-107.101 21.162-107.101 21.743Q-107.101 22.314-106.867 22.704Q-106.633 23.093-106.093 23.093M-103.393 21.764Q-103.393 21.422-103.258 21.123Q-103.123 20.824-102.883 20.600Q-102.644 20.376-102.326 20.251Q-102.008 20.126-101.677 20.126Q-101.232 20.126-100.832 20.342Q-100.433 20.557-100.198 20.935Q-99.964 21.312-99.964 21.764Q-99.964 22.105-100.106 22.389Q-100.248 22.673-100.492 22.880Q-100.737 23.086-101.046 23.201Q-101.355 23.315-101.677 23.315Q-102.107 23.315-102.509 23.114Q-102.911 22.912-103.152 22.560Q-103.393 22.208-103.393 21.764M-101.677 23.066Q-101.075 23.066-100.851 22.688Q-100.627 22.310-100.627 21.678Q-100.627 21.066-100.862 20.707Q-101.096 20.349-101.677 20.349Q-102.729 20.349-102.729 21.678Q-102.729 22.310-102.504 22.688Q-102.278 23.066-101.677 23.066M-97.620 23.247L-99.356 23.247L-99.356 22.967Q-99.127 22.967-98.978 22.933Q-98.830 22.898-98.830 22.758L-98.830 20.909Q-98.830 20.639-98.937 20.578Q-99.045 20.516-99.356 20.516L-99.356 20.236L-98.327 20.161L-98.327 20.868Q-98.197 20.560-97.955 20.361Q-97.712 20.161-97.394 20.161Q-97.175 20.161-97.004 20.285Q-96.833 20.410-96.833 20.622Q-96.833 20.759-96.933 20.858Q-97.032 20.957-97.165 20.957Q-97.302 20.957-97.401 20.858Q-97.500 20.759-97.500 20.622Q-97.500 20.482-97.401 20.383Q-97.691 20.383-97.891 20.579Q-98.091 20.776-98.184 21.070Q-98.276 21.364-98.276 21.644L-98.276 22.758Q-98.276 22.967-97.620 22.967L-97.620 23.247M-95.723 22.406L-95.723 20.509L-96.362 20.509L-96.362 20.287Q-96.044 20.287-95.827 20.077Q-95.610 19.867-95.509 19.557Q-95.408 19.248-95.408 18.940L-95.142 18.940L-95.142 20.229L-94.065 20.229L-94.065 20.509L-95.142 20.509L-95.142 22.393Q-95.142 22.669-95.037 22.868Q-94.933 23.066-94.673 23.066Q-94.516 23.066-94.410 22.962Q-94.304 22.857-94.255 22.704Q-94.205 22.550-94.205 22.393L-94.205 21.979L-93.938 21.979L-93.938 22.406Q-93.938 22.632-94.038 22.842Q-94.137 23.052-94.321 23.184Q-94.506 23.315-94.735 23.315Q-95.172 23.315-95.447 23.078Q-95.723 22.840-95.723 22.406\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M48.408-9.474H76.86v-19.917H48.408Z\"\u002F>\u003Cg transform=\"translate(164.953 -41.172)\">\u003Cpath d=\"M-107.809 21.712Q-107.809 21.391-107.684 21.102Q-107.559 20.813-107.333 20.590Q-107.108 20.366-106.812 20.246Q-106.517 20.126-106.199 20.126Q-105.871 20.126-105.609 20.226Q-105.348 20.325-105.172 20.507Q-104.996 20.690-104.902 20.948Q-104.808 21.206-104.808 21.538Q-104.808 21.630-104.890 21.651L-107.145 21.651L-107.145 21.712Q-107.145 22.300-106.862 22.683Q-106.578 23.066-106.011 23.066Q-105.689 23.066-105.421 22.873Q-105.153 22.680-105.064 22.365Q-105.057 22.324-104.982 22.310L-104.890 22.310Q-104.808 22.334-104.808 22.406Q-104.808 22.413-104.814 22.440Q-104.927 22.837-105.298 23.076Q-105.669 23.315-106.093 23.315Q-106.530 23.315-106.930 23.107Q-107.330 22.898-107.569 22.531Q-107.809 22.164-107.809 21.712M-107.139 21.442L-105.324 21.442Q-105.324 21.165-105.421 20.913Q-105.519 20.660-105.717 20.504Q-105.915 20.349-106.199 20.349Q-106.476 20.349-106.689 20.507Q-106.903 20.666-107.021 20.921Q-107.139 21.176-107.139 21.442M-102.538 23.247L-104.172 23.247L-104.172 22.967Q-103.943 22.967-103.794 22.933Q-103.645 22.898-103.645 22.758L-103.645 20.909Q-103.645 20.639-103.753 20.578Q-103.861 20.516-104.172 20.516L-104.172 20.236L-103.112 20.161L-103.112 20.810Q-102.941 20.502-102.637 20.331Q-102.333 20.161-101.988 20.161Q-101.482 20.161-101.198 20.384Q-100.915 20.608-100.915 21.104L-100.915 22.758Q-100.915 22.895-100.766 22.931Q-100.617 22.967-100.392 22.967L-100.392 23.247L-102.022 23.247L-102.022 22.967Q-101.793 22.967-101.644 22.933Q-101.496 22.898-101.496 22.758L-101.496 21.118Q-101.496 20.783-101.615 20.583Q-101.735 20.383-102.049 20.383Q-102.319 20.383-102.553 20.519Q-102.788 20.656-102.926 20.890Q-103.064 21.124-103.064 21.398L-103.064 22.758Q-103.064 22.895-102.914 22.931Q-102.764 22.967-102.538 22.967L-102.538 23.247M-99.804 21.736Q-99.804 21.408-99.669 21.107Q-99.534 20.807-99.298 20.586Q-99.062 20.366-98.758 20.246Q-98.454 20.126-98.129 20.126Q-97.623 20.126-97.274 20.229Q-96.926 20.331-96.926 20.707Q-96.926 20.854-97.023 20.955Q-97.121 21.056-97.268 21.056Q-97.421 21.056-97.520 20.957Q-97.620 20.858-97.620 20.707Q-97.620 20.519-97.479 20.427Q-97.681 20.376-98.122 20.376Q-98.478 20.376-98.707 20.572Q-98.936 20.769-99.036 21.078Q-99.137 21.388-99.137 21.736Q-99.137 22.085-99.011 22.391Q-98.884 22.697-98.630 22.881Q-98.375 23.066-98.019 23.066Q-97.797 23.066-97.613 22.982Q-97.428 22.898-97.293 22.743Q-97.158 22.587-97.100 22.379Q-97.086 22.324-97.032 22.324L-96.919 22.324Q-96.888 22.324-96.866 22.348Q-96.844 22.372-96.844 22.406L-96.844 22.427Q-96.929 22.714-97.117 22.912Q-97.305 23.110-97.570 23.213Q-97.835 23.315-98.129 23.315Q-98.560 23.315-98.947 23.109Q-99.335 22.902-99.570 22.539Q-99.804 22.177-99.804 21.736\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M62.634 13.088V-6.674\"\u002F>\u003Cpath stroke=\"none\" d=\"m62.634-9.274-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M-36.75-19.432h25.452\"\u002F>\u003Cpath stroke=\"none\" d=\"m-8.698-19.432-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M20.155-19.432h25.453\"\u002F>\u003Cpath stroke=\"none\" d=\"m48.208-19.432-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M109.58-9.474h31.299v-19.917H109.58Z\"\u002F>\u003Cg transform=\"translate(231.535 -41.172)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.408-107.633 21.107Q-107.498 20.807-107.262 20.586Q-107.026 20.366-106.722 20.246Q-106.417 20.126-106.093 20.126Q-105.587 20.126-105.238 20.229Q-104.890 20.331-104.890 20.707Q-104.890 20.854-104.987 20.955Q-105.084 21.056-105.231 21.056Q-105.385 21.056-105.484 20.957Q-105.583 20.858-105.583 20.707Q-105.583 20.519-105.443 20.427Q-105.645 20.376-106.086 20.376Q-106.441 20.376-106.670 20.572Q-106.899 20.769-107 21.078Q-107.101 21.388-107.101 21.736Q-107.101 22.085-106.975 22.391Q-106.848 22.697-106.593 22.881Q-106.339 23.066-105.983 23.066Q-105.761 23.066-105.577 22.982Q-105.392 22.898-105.257 22.743Q-105.122 22.587-105.064 22.379Q-105.050 22.324-104.996 22.324L-104.883 22.324Q-104.852 22.324-104.830 22.348Q-104.808 22.372-104.808 22.406L-104.808 22.427Q-104.893 22.714-105.081 22.912Q-105.269 23.110-105.534 23.213Q-105.799 23.315-106.093 23.315Q-106.523 23.315-106.911 23.109Q-107.299 22.902-107.533 22.539Q-107.768 22.177-107.768 21.736\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M77.06-19.432h29.521\"\u002F>\u003Cpath stroke=\"none\" d=\"m109.18-19.432-4.159-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003Cg fill=\"var(--tk-accent)\" stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(188.328 -46.412)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.408-107.633 21.107Q-107.498 20.807-107.262 20.586Q-107.026 20.366-106.722 20.246Q-106.417 20.126-106.093 20.126Q-105.587 20.126-105.238 20.229Q-104.890 20.331-104.890 20.707Q-104.890 20.854-104.987 20.955Q-105.084 21.056-105.231 21.056Q-105.385 21.056-105.484 20.957Q-105.583 20.858-105.583 20.707Q-105.583 20.519-105.443 20.427Q-105.645 20.376-106.086 20.376Q-106.441 20.376-106.670 20.572Q-106.899 20.769-107 21.078Q-107.101 21.388-107.101 21.736Q-107.101 22.085-106.975 22.391Q-106.848 22.697-106.593 22.881Q-106.339 23.066-105.983 23.066Q-105.761 23.066-105.577 22.982Q-105.392 22.898-105.257 22.743Q-105.122 22.587-105.064 22.379Q-105.050 22.324-104.996 22.324L-104.883 22.324Q-104.852 22.324-104.830 22.348Q-104.808 22.372-104.808 22.406L-104.808 22.427Q-104.893 22.714-105.081 22.912Q-105.269 23.110-105.534 23.213Q-105.799 23.315-106.093 23.315Q-106.523 23.315-106.911 23.109Q-107.299 22.902-107.533 22.539Q-107.768 22.177-107.768 21.736M-104.261 21.764Q-104.261 21.422-104.126 21.123Q-103.991 20.824-103.751 20.600Q-103.512 20.376-103.194 20.251Q-102.876 20.126-102.545 20.126Q-102.101 20.126-101.701 20.342Q-101.301 20.557-101.067 20.935Q-100.832 21.312-100.832 21.764Q-100.832 22.105-100.974 22.389Q-101.116 22.673-101.361 22.880Q-101.605 23.086-101.914 23.201Q-102.224 23.315-102.545 23.315Q-102.976 23.315-103.377 23.114Q-103.779 22.912-104.020 22.560Q-104.261 22.208-104.261 21.764M-102.545 23.066Q-101.943 23.066-101.719 22.688Q-101.496 22.310-101.496 21.678Q-101.496 21.066-101.730 20.707Q-101.964 20.349-102.545 20.349Q-103.598 20.349-103.598 21.678Q-103.598 22.310-103.372 22.688Q-103.146 23.066-102.545 23.066M-98.556 23.247L-100.190 23.247L-100.190 22.967Q-99.961 22.967-99.812 22.933Q-99.664 22.898-99.664 22.758L-99.664 20.909Q-99.664 20.639-99.771 20.578Q-99.879 20.516-100.190 20.516L-100.190 20.236L-99.130 20.161L-99.130 20.810Q-98.959 20.502-98.655 20.331Q-98.351 20.161-98.006 20.161Q-97.500 20.161-97.216 20.384Q-96.933 20.608-96.933 21.104L-96.933 22.758Q-96.933 22.895-96.784 22.931Q-96.635 22.967-96.410 22.967L-96.410 23.247L-98.040 23.247L-98.040 22.967Q-97.811 22.967-97.662 22.933Q-97.514 22.898-97.514 22.758L-97.514 21.118Q-97.514 20.783-97.633 20.583Q-97.753 20.383-98.067 20.383Q-98.337 20.383-98.572 20.519Q-98.806 20.656-98.944 20.890Q-99.082 21.124-99.082 21.398L-99.082 22.758Q-99.082 22.895-98.932 22.931Q-98.782 22.967-98.556 22.967\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(188.328 -46.412)\">\u003Cpath d=\"M-95.498 22.406L-95.498 20.509L-96.137 20.509L-96.137 20.287Q-95.819 20.287-95.602 20.077Q-95.385 19.867-95.285 19.557Q-95.184 19.248-95.184 18.940L-94.917 18.940L-94.917 20.229L-93.840 20.229L-93.840 20.509L-94.917 20.509L-94.917 22.393Q-94.917 22.669-94.813 22.868Q-94.709 23.066-94.449 23.066Q-94.292 23.066-94.186 22.962Q-94.080 22.857-94.030 22.704Q-93.981 22.550-93.981 22.393L-93.981 21.979L-93.714 21.979L-93.714 22.406Q-93.714 22.632-93.813 22.842Q-93.912 23.052-94.097 23.184Q-94.281 23.315-94.510 23.315Q-94.948 23.315-95.223 23.078Q-95.498 22.840-95.498 22.406M-92.945 21.712Q-92.945 21.391-92.820 21.102Q-92.695 20.813-92.470 20.590Q-92.244 20.366-91.949 20.246Q-91.653 20.126-91.335 20.126Q-91.007 20.126-90.745 20.226Q-90.484 20.325-90.308 20.507Q-90.132 20.690-90.038 20.948Q-89.944 21.206-89.944 21.538Q-89.944 21.630-90.026 21.651L-92.282 21.651L-92.282 21.712Q-92.282 22.300-91.998 22.683Q-91.714 23.066-91.147 23.066Q-90.826 23.066-90.558 22.873Q-90.289 22.680-90.200 22.365Q-90.193 22.324-90.118 22.310L-90.026 22.310Q-89.944 22.334-89.944 22.406Q-89.944 22.413-89.951 22.440Q-90.064 22.837-90.434 23.076Q-90.805 23.315-91.229 23.315Q-91.667 23.315-92.067 23.107Q-92.466 22.898-92.706 22.531Q-92.945 22.164-92.945 21.712M-92.275 21.442L-90.460 21.442Q-90.460 21.165-90.558 20.913Q-90.655 20.660-90.853 20.504Q-91.051 20.349-91.335 20.349Q-91.612 20.349-91.826 20.507Q-92.039 20.666-92.157 20.921Q-92.275 21.176-92.275 21.442M-88.173 23.247L-89.496 23.247L-89.496 22.967Q-88.936 22.967-88.556 22.567L-87.842 21.770L-88.755 20.721Q-88.891 20.574-89.040 20.542Q-89.189 20.509-89.455 20.509L-89.455 20.229L-87.955 20.229L-87.955 20.509Q-88.146 20.509-88.146 20.643Q-88.146 20.673-88.115 20.721L-87.521 21.405L-87.080 20.909Q-86.967 20.779-86.967 20.663Q-86.967 20.601-87.005 20.555Q-87.042 20.509-87.100 20.509L-87.100 20.229L-85.784 20.229L-85.784 20.509Q-86.345 20.509-86.724 20.909L-87.346 21.610L-86.352 22.758Q-86.253 22.857-86.152 22.902Q-86.051 22.946-85.940 22.956Q-85.829 22.967-85.651 22.967L-85.651 23.247L-87.145 23.247L-87.145 22.967Q-87.080 22.967-87.020 22.933Q-86.960 22.898-86.960 22.833Q-86.960 22.786-86.991 22.758L-87.668 21.972L-88.201 22.567Q-88.314 22.697-88.314 22.813Q-88.314 22.878-88.273 22.922Q-88.232 22.967-88.173 22.967L-88.173 23.247M-84.629 22.406L-84.629 20.509L-85.268 20.509L-85.268 20.287Q-84.950 20.287-84.733 20.077Q-84.516 19.867-84.415 19.557Q-84.315 19.248-84.315 18.940L-84.048 18.940L-84.048 20.229L-82.971 20.229L-82.971 20.509L-84.048 20.509L-84.048 22.393Q-84.048 22.669-83.944 22.868Q-83.839 23.066-83.580 23.066Q-83.422 23.066-83.317 22.962Q-83.211 22.857-83.161 22.704Q-83.111 22.550-83.111 22.393L-83.111 21.979L-82.845 21.979L-82.845 22.406Q-82.845 22.632-82.944 22.842Q-83.043 23.052-83.228 23.184Q-83.412 23.315-83.641 23.315Q-84.079 23.315-84.354 23.078Q-84.629 22.840-84.629 22.406\" fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M167.91-9.474h28.452v-19.917h-28.453Z\"\u002F>\u003Cg transform=\"translate(284.454 -40.249)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.398-107.627 21.107Q-107.487 20.817-107.243 20.603Q-106.999 20.390-106.694 20.275Q-106.390 20.161-106.065 20.161Q-105.795 20.161-105.532 20.260Q-105.269 20.359-105.078 20.537L-105.078 19.139Q-105.078 18.869-105.185 18.807Q-105.293 18.746-105.604 18.746L-105.604 18.465L-104.527 18.390L-104.527 22.574Q-104.527 22.762-104.473 22.845Q-104.418 22.929-104.317 22.948Q-104.216 22.967-104.001 22.967L-104.001 23.247L-105.108 23.315L-105.108 22.898Q-105.525 23.315-106.151 23.315Q-106.582 23.315-106.954 23.103Q-107.327 22.892-107.547 22.531Q-107.768 22.170-107.768 21.736M-106.093 23.093Q-105.884 23.093-105.698 23.021Q-105.512 22.950-105.358 22.813Q-105.204 22.676-105.108 22.498L-105.108 20.889Q-105.194 20.742-105.339 20.622Q-105.484 20.502-105.654 20.443Q-105.823 20.383-106.004 20.383Q-106.564 20.383-106.833 20.772Q-107.101 21.162-107.101 21.743Q-107.101 22.314-106.867 22.704Q-106.633 23.093-106.093 23.093M-103.393 21.712Q-103.393 21.391-103.268 21.102Q-103.143 20.813-102.917 20.590Q-102.692 20.366-102.396 20.246Q-102.101 20.126-101.783 20.126Q-101.455 20.126-101.193 20.226Q-100.932 20.325-100.756 20.507Q-100.580 20.690-100.486 20.948Q-100.392 21.206-100.392 21.538Q-100.392 21.630-100.474 21.651L-102.729 21.651L-102.729 21.712Q-102.729 22.300-102.446 22.683Q-102.162 23.066-101.595 23.066Q-101.273 23.066-101.005 22.873Q-100.737 22.680-100.648 22.365Q-100.641 22.324-100.566 22.310L-100.474 22.310Q-100.392 22.334-100.392 22.406Q-100.392 22.413-100.398 22.440Q-100.511 22.837-100.882 23.076Q-101.253 23.315-101.677 23.315Q-102.114 23.315-102.514 23.107Q-102.914 22.898-103.153 22.531Q-103.393 22.164-103.393 21.712M-102.723 21.442L-100.908 21.442Q-100.908 21.165-101.005 20.913Q-101.103 20.660-101.301 20.504Q-101.499 20.349-101.783 20.349Q-102.060 20.349-102.273 20.507Q-102.487 20.666-102.605 20.921Q-102.723 21.176-102.723 21.442M-99.804 21.736Q-99.804 21.408-99.669 21.107Q-99.534 20.807-99.298 20.586Q-99.062 20.366-98.758 20.246Q-98.454 20.126-98.129 20.126Q-97.623 20.126-97.274 20.229Q-96.926 20.331-96.926 20.707Q-96.926 20.854-97.023 20.955Q-97.121 21.056-97.268 21.056Q-97.421 21.056-97.520 20.957Q-97.620 20.858-97.620 20.707Q-97.620 20.519-97.479 20.427Q-97.681 20.376-98.122 20.376Q-98.478 20.376-98.707 20.572Q-98.936 20.769-99.036 21.078Q-99.137 21.388-99.137 21.736Q-99.137 22.085-99.011 22.391Q-98.884 22.697-98.630 22.881Q-98.375 23.066-98.019 23.066Q-97.797 23.066-97.613 22.982Q-97.428 22.898-97.293 22.743Q-97.158 22.587-97.100 22.379Q-97.086 22.324-97.032 22.324L-96.919 22.324Q-96.888 22.324-96.866 22.348Q-96.844 22.372-96.844 22.406L-96.844 22.427Q-96.929 22.714-97.117 22.912Q-97.305 23.110-97.570 23.213Q-97.835 23.315-98.129 23.315Q-98.560 23.315-98.947 23.109Q-99.335 22.902-99.570 22.539Q-99.804 22.177-99.804 21.736\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M167.91 33.205h28.452V13.288h-28.453Z\"\u002F>\u003Cg transform=\"translate(281.953 2.153)\">\u003Cpath d=\"M-107.768 23.240L-107.768 22.177Q-107.768 22.153-107.740 22.126Q-107.713 22.099-107.689 22.099L-107.580 22.099Q-107.515 22.099-107.501 22.157Q-107.405 22.591-107.159 22.842Q-106.913 23.093-106.499 23.093Q-106.158 23.093-105.905 22.960Q-105.652 22.827-105.652 22.519Q-105.652 22.362-105.746 22.247Q-105.840 22.133-105.978 22.064Q-106.117 21.996-106.284 21.958L-106.865 21.859Q-107.221 21.791-107.494 21.570Q-107.768 21.350-107.768 21.008Q-107.768 20.759-107.656 20.584Q-107.545 20.410-107.359 20.311Q-107.173 20.212-106.957 20.169Q-106.742 20.126-106.499 20.126Q-106.086 20.126-105.806 20.308L-105.590 20.133Q-105.580 20.130-105.573 20.128Q-105.566 20.126-105.556 20.126L-105.505 20.126Q-105.478 20.126-105.454 20.150Q-105.430 20.174-105.430 20.202L-105.430 21.049Q-105.430 21.070-105.454 21.097Q-105.478 21.124-105.505 21.124L-105.618 21.124Q-105.645 21.124-105.671 21.099Q-105.696 21.073-105.696 21.049Q-105.696 20.813-105.802 20.649Q-105.908 20.485-106.091 20.403Q-106.274 20.321-106.506 20.321Q-106.834 20.321-107.091 20.424Q-107.347 20.526-107.347 20.803Q-107.347 20.998-107.164 21.107Q-106.981 21.217-106.752 21.258L-106.178 21.364Q-105.932 21.412-105.718 21.540Q-105.505 21.668-105.368 21.871Q-105.231 22.075-105.231 22.324Q-105.231 22.837-105.597 23.076Q-105.963 23.315-106.499 23.315Q-106.995 23.315-107.327 23.021L-107.593 23.295Q-107.614 23.315-107.641 23.315L-107.689 23.315Q-107.713 23.315-107.740 23.288Q-107.768 23.261-107.768 23.240M-104.076 22.406L-104.076 20.509L-104.715 20.509L-104.715 20.287Q-104.397 20.287-104.180 20.077Q-103.963 19.867-103.863 19.557Q-103.762 19.248-103.762 18.940L-103.495 18.940L-103.495 20.229L-102.418 20.229L-102.418 20.509L-103.495 20.509L-103.495 22.393Q-103.495 22.669-103.391 22.868Q-103.287 23.066-103.027 23.066Q-102.870 23.066-102.764 22.962Q-102.658 22.857-102.608 22.704Q-102.559 22.550-102.559 22.393L-102.559 21.979L-102.292 21.979L-102.292 22.406Q-102.292 22.632-102.391 22.842Q-102.490 23.052-102.675 23.184Q-102.859 23.315-103.088 23.315Q-103.526 23.315-103.801 23.078Q-104.076 22.840-104.076 22.406M-101.424 22.519Q-101.424 22.187-101.200 21.960Q-100.976 21.733-100.633 21.605Q-100.289 21.476-99.916 21.424Q-99.544 21.371-99.240 21.371L-99.240 21.118Q-99.240 20.913-99.347 20.733Q-99.455 20.554-99.636 20.451Q-99.817 20.349-100.026 20.349Q-100.433 20.349-100.668 20.441Q-100.580 20.478-100.533 20.562Q-100.487 20.646-100.487 20.748Q-100.487 20.844-100.533 20.923Q-100.580 21.001-100.660 21.046Q-100.740 21.090-100.829 21.090Q-100.979 21.090-101.080 20.993Q-101.181 20.895-101.181 20.748Q-101.181 20.126-100.026 20.126Q-99.814 20.126-99.564 20.190Q-99.315 20.253-99.113 20.372Q-98.912 20.492-98.785 20.677Q-98.659 20.861-98.659 21.104L-98.659 22.680Q-98.659 22.796-98.597 22.892Q-98.536 22.987-98.423 22.987Q-98.313 22.987-98.249 22.893Q-98.184 22.799-98.184 22.680L-98.184 22.232L-97.917 22.232L-97.917 22.680Q-97.917 22.950-98.144 23.115Q-98.372 23.281-98.652 23.281Q-98.860 23.281-98.997 23.127Q-99.134 22.974-99.158 22.758Q-99.305 23.025-99.587 23.170Q-99.869 23.315-100.193 23.315Q-100.470 23.315-100.754 23.240Q-101.038 23.165-101.231 22.986Q-101.424 22.806-101.424 22.519M-100.809 22.519Q-100.809 22.693-100.708 22.823Q-100.607 22.953-100.451 23.023Q-100.296 23.093-100.132 23.093Q-99.913 23.093-99.705 22.996Q-99.496 22.898-99.368 22.717Q-99.240 22.536-99.240 22.310L-99.240 21.582Q-99.564 21.582-99.930 21.673Q-100.296 21.764-100.552 21.976Q-100.809 22.187-100.809 22.519M-95.750 23.247L-97.486 23.247L-97.486 22.967Q-97.257 22.967-97.109 22.933Q-96.960 22.898-96.960 22.758L-96.960 20.909Q-96.960 20.639-97.068 20.578Q-97.175 20.516-97.486 20.516L-97.486 20.236L-96.457 20.161L-96.457 20.868Q-96.328 20.560-96.085 20.361Q-95.842 20.161-95.524 20.161Q-95.306 20.161-95.135 20.285Q-94.964 20.410-94.964 20.622Q-94.964 20.759-95.063 20.858Q-95.162 20.957-95.295 20.957Q-95.432 20.957-95.531 20.858Q-95.630 20.759-95.630 20.622Q-95.630 20.482-95.531 20.383Q-95.822 20.383-96.022 20.579Q-96.222 20.776-96.314 21.070Q-96.406 21.364-96.406 21.644L-96.406 22.758Q-96.406 22.967-95.750 22.967L-95.750 23.247M-93.853 22.406L-93.853 20.509L-94.492 20.509L-94.492 20.287Q-94.174 20.287-93.957 20.077Q-93.740 19.867-93.639 19.557Q-93.539 19.248-93.539 18.940L-93.272 18.940L-93.272 20.229L-92.195 20.229L-92.195 20.509L-93.272 20.509L-93.272 22.393Q-93.272 22.669-93.168 22.868Q-93.063 23.066-92.804 23.066Q-92.646 23.066-92.540 22.962Q-92.435 22.857-92.385 22.704Q-92.335 22.550-92.335 22.393L-92.335 21.979L-92.069 21.979L-92.069 22.406Q-92.069 22.632-92.168 22.842Q-92.267 23.052-92.452 23.184Q-92.636 23.315-92.865 23.315Q-93.303 23.315-93.578 23.078Q-93.853 22.840-93.853 22.406\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M167.91-52.153h28.452V-72.07h-28.453Z\"\u002F>\u003Cg transform=\"translate(284.67 -82.928)\">\u003Cpath d=\"M-107.241 22.406L-107.241 20.509L-107.880 20.509L-107.880 20.287Q-107.562 20.287-107.345 20.077Q-107.128 19.867-107.028 19.557Q-106.927 19.248-106.927 18.940L-106.660 18.940L-106.660 20.229L-105.583 20.229L-105.583 20.509L-106.660 20.509L-106.660 22.393Q-106.660 22.669-106.556 22.868Q-106.452 23.066-106.192 23.066Q-106.035 23.066-105.929 22.962Q-105.823 22.857-105.773 22.704Q-105.724 22.550-105.724 22.393L-105.724 21.979L-105.457 21.979L-105.457 22.406Q-105.457 22.632-105.556 22.842Q-105.655 23.052-105.840 23.184Q-106.024 23.315-106.253 23.315Q-106.691 23.315-106.966 23.078Q-107.241 22.840-107.241 22.406M-102.965 23.247L-104.599 23.247L-104.599 22.967Q-104.370 22.967-104.221 22.933Q-104.073 22.898-104.073 22.758L-104.073 19.139Q-104.073 18.869-104.180 18.807Q-104.288 18.746-104.599 18.746L-104.599 18.465L-103.519 18.390L-103.519 20.776Q-103.413 20.591-103.235 20.449Q-103.058 20.308-102.849 20.234Q-102.641 20.161-102.415 20.161Q-101.909 20.161-101.625 20.384Q-101.342 20.608-101.342 21.104L-101.342 22.758Q-101.342 22.895-101.193 22.931Q-101.044 22.967-100.819 22.967L-100.819 23.247L-102.449 23.247L-102.449 22.967Q-102.220 22.967-102.072 22.933Q-101.923 22.898-101.923 22.758L-101.923 21.118Q-101.923 20.783-102.042 20.583Q-102.162 20.383-102.477 20.383Q-102.747 20.383-102.981 20.519Q-103.215 20.656-103.353 20.890Q-103.492 21.124-103.492 21.398L-103.492 22.758Q-103.492 22.895-103.341 22.931Q-103.191 22.967-102.965 22.967L-102.965 23.247M-100.272 21.712Q-100.272 21.391-100.147 21.102Q-100.022 20.813-99.797 20.590Q-99.571 20.366-99.276 20.246Q-98.980 20.126-98.662 20.126Q-98.334 20.126-98.072 20.226Q-97.811 20.325-97.635 20.507Q-97.459 20.690-97.365 20.948Q-97.271 21.206-97.271 21.538Q-97.271 21.630-97.353 21.651L-99.609 21.651L-99.609 21.712Q-99.609 22.300-99.325 22.683Q-99.041 23.066-98.474 23.066Q-98.153 23.066-97.884 22.873Q-97.616 22.680-97.527 22.365Q-97.520 22.324-97.445 22.310L-97.353 22.310Q-97.271 22.334-97.271 22.406Q-97.271 22.413-97.278 22.440Q-97.391 22.837-97.761 23.076Q-98.132 23.315-98.556 23.315Q-98.994 23.315-99.394 23.107Q-99.793 22.898-100.033 22.531Q-100.272 22.164-100.272 21.712M-99.602 21.442L-97.787 21.442Q-97.787 21.165-97.884 20.913Q-97.982 20.660-98.180 20.504Q-98.378 20.349-98.662 20.349Q-98.939 20.349-99.153 20.507Q-99.366 20.666-99.484 20.921Q-99.602 21.176-99.602 21.442\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M182.135 13.088V-6.674\"\u002F>\u003Cpath stroke=\"none\" d=\"m182.135-9.274-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M182.135-29.59v-19.763\"\u002F>\u003Cpath stroke=\"none\" d=\"m182.135-51.953-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M224.815-9.474h28.452v-19.917h-28.452Z\"\u002F>\u003Cg transform=\"translate(341.36 -40.249)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.398-107.627 21.107Q-107.487 20.817-107.243 20.603Q-106.999 20.390-106.694 20.275Q-106.390 20.161-106.065 20.161Q-105.795 20.161-105.532 20.260Q-105.269 20.359-105.078 20.537L-105.078 19.139Q-105.078 18.869-105.185 18.807Q-105.293 18.746-105.604 18.746L-105.604 18.465L-104.527 18.390L-104.527 22.574Q-104.527 22.762-104.473 22.845Q-104.418 22.929-104.317 22.948Q-104.216 22.967-104.001 22.967L-104.001 23.247L-105.108 23.315L-105.108 22.898Q-105.525 23.315-106.151 23.315Q-106.582 23.315-106.954 23.103Q-107.327 22.892-107.547 22.531Q-107.768 22.170-107.768 21.736M-106.093 23.093Q-105.884 23.093-105.698 23.021Q-105.512 22.950-105.358 22.813Q-105.204 22.676-105.108 22.498L-105.108 20.889Q-105.194 20.742-105.339 20.622Q-105.484 20.502-105.654 20.443Q-105.823 20.383-106.004 20.383Q-106.564 20.383-106.833 20.772Q-107.101 21.162-107.101 21.743Q-107.101 22.314-106.867 22.704Q-106.633 23.093-106.093 23.093M-103.393 21.712Q-103.393 21.391-103.268 21.102Q-103.143 20.813-102.917 20.590Q-102.692 20.366-102.396 20.246Q-102.101 20.126-101.783 20.126Q-101.455 20.126-101.193 20.226Q-100.932 20.325-100.756 20.507Q-100.580 20.690-100.486 20.948Q-100.392 21.206-100.392 21.538Q-100.392 21.630-100.474 21.651L-102.729 21.651L-102.729 21.712Q-102.729 22.300-102.446 22.683Q-102.162 23.066-101.595 23.066Q-101.273 23.066-101.005 22.873Q-100.737 22.680-100.648 22.365Q-100.641 22.324-100.566 22.310L-100.474 22.310Q-100.392 22.334-100.392 22.406Q-100.392 22.413-100.398 22.440Q-100.511 22.837-100.882 23.076Q-101.253 23.315-101.677 23.315Q-102.114 23.315-102.514 23.107Q-102.914 22.898-103.153 22.531Q-103.393 22.164-103.393 21.712M-102.723 21.442L-100.908 21.442Q-100.908 21.165-101.005 20.913Q-101.103 20.660-101.301 20.504Q-101.499 20.349-101.783 20.349Q-102.060 20.349-102.273 20.507Q-102.487 20.666-102.605 20.921Q-102.723 21.176-102.723 21.442M-99.804 21.736Q-99.804 21.408-99.669 21.107Q-99.534 20.807-99.298 20.586Q-99.062 20.366-98.758 20.246Q-98.454 20.126-98.129 20.126Q-97.623 20.126-97.274 20.229Q-96.926 20.331-96.926 20.707Q-96.926 20.854-97.023 20.955Q-97.121 21.056-97.268 21.056Q-97.421 21.056-97.520 20.957Q-97.620 20.858-97.620 20.707Q-97.620 20.519-97.479 20.427Q-97.681 20.376-98.122 20.376Q-98.478 20.376-98.707 20.572Q-98.936 20.769-99.036 21.078Q-99.137 21.388-99.137 21.736Q-99.137 22.085-99.011 22.391Q-98.884 22.697-98.630 22.881Q-98.375 23.066-98.019 23.066Q-97.797 23.066-97.613 22.982Q-97.428 22.898-97.293 22.743Q-97.158 22.587-97.100 22.379Q-97.086 22.324-97.032 22.324L-96.919 22.324Q-96.888 22.324-96.866 22.348Q-96.844 22.372-96.844 22.406L-96.844 22.427Q-96.929 22.714-97.117 22.912Q-97.305 23.110-97.570 23.213Q-97.835 23.315-98.129 23.315Q-98.560 23.315-98.947 23.109Q-99.335 22.902-99.570 22.539Q-99.804 22.177-99.804 21.736\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M224.815 33.205h28.452V13.288h-28.452Z\"\u002F>\u003Cg transform=\"translate(341.575 2.43)\">\u003Cpath d=\"M-107.241 22.406L-107.241 20.509L-107.880 20.509L-107.880 20.287Q-107.562 20.287-107.345 20.077Q-107.128 19.867-107.028 19.557Q-106.927 19.248-106.927 18.940L-106.660 18.940L-106.660 20.229L-105.583 20.229L-105.583 20.509L-106.660 20.509L-106.660 22.393Q-106.660 22.669-106.556 22.868Q-106.452 23.066-106.192 23.066Q-106.035 23.066-105.929 22.962Q-105.823 22.857-105.773 22.704Q-105.724 22.550-105.724 22.393L-105.724 21.979L-105.457 21.979L-105.457 22.406Q-105.457 22.632-105.556 22.842Q-105.655 23.052-105.840 23.184Q-106.024 23.315-106.253 23.315Q-106.691 23.315-106.966 23.078Q-107.241 22.840-107.241 22.406M-102.965 23.247L-104.599 23.247L-104.599 22.967Q-104.370 22.967-104.221 22.933Q-104.073 22.898-104.073 22.758L-104.073 19.139Q-104.073 18.869-104.180 18.807Q-104.288 18.746-104.599 18.746L-104.599 18.465L-103.519 18.390L-103.519 20.776Q-103.413 20.591-103.235 20.449Q-103.058 20.308-102.849 20.234Q-102.641 20.161-102.415 20.161Q-101.909 20.161-101.625 20.384Q-101.342 20.608-101.342 21.104L-101.342 22.758Q-101.342 22.895-101.193 22.931Q-101.044 22.967-100.819 22.967L-100.819 23.247L-102.449 23.247L-102.449 22.967Q-102.220 22.967-102.072 22.933Q-101.923 22.898-101.923 22.758L-101.923 21.118Q-101.923 20.783-102.042 20.583Q-102.162 20.383-102.477 20.383Q-102.747 20.383-102.981 20.519Q-103.215 20.656-103.353 20.890Q-103.492 21.124-103.492 21.398L-103.492 22.758Q-103.492 22.895-103.341 22.931Q-103.191 22.967-102.965 22.967L-102.965 23.247M-100.272 21.712Q-100.272 21.391-100.147 21.102Q-100.022 20.813-99.797 20.590Q-99.571 20.366-99.276 20.246Q-98.980 20.126-98.662 20.126Q-98.334 20.126-98.072 20.226Q-97.811 20.325-97.635 20.507Q-97.459 20.690-97.365 20.948Q-97.271 21.206-97.271 21.538Q-97.271 21.630-97.353 21.651L-99.609 21.651L-99.609 21.712Q-99.609 22.300-99.325 22.683Q-99.041 23.066-98.474 23.066Q-98.153 23.066-97.884 22.873Q-97.616 22.680-97.527 22.365Q-97.520 22.324-97.445 22.310L-97.353 22.310Q-97.271 22.334-97.271 22.406Q-97.271 22.413-97.278 22.440Q-97.391 22.837-97.761 23.076Q-98.132 23.315-98.556 23.315Q-98.994 23.315-99.394 23.107Q-99.793 22.898-100.033 22.531Q-100.272 22.164-100.272 21.712M-99.602 21.442L-97.787 21.442Q-97.787 21.165-97.884 20.913Q-97.982 20.660-98.180 20.504Q-98.378 20.349-98.662 20.349Q-98.939 20.349-99.153 20.507Q-99.366 20.666-99.484 20.921Q-99.602 21.176-99.602 21.442\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M224.815-52.153h28.452V-72.07h-28.452Z\"\u002F>\u003Cg transform=\"translate(341.79 -83.205)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.408-107.633 21.107Q-107.498 20.807-107.262 20.586Q-107.026 20.366-106.722 20.246Q-106.417 20.126-106.093 20.126Q-105.587 20.126-105.238 20.229Q-104.890 20.331-104.890 20.707Q-104.890 20.854-104.987 20.955Q-105.084 21.056-105.231 21.056Q-105.385 21.056-105.484 20.957Q-105.583 20.858-105.583 20.707Q-105.583 20.519-105.443 20.427Q-105.645 20.376-106.086 20.376Q-106.441 20.376-106.670 20.572Q-106.899 20.769-107 21.078Q-107.101 21.388-107.101 21.736Q-107.101 22.085-106.975 22.391Q-106.848 22.697-106.593 22.881Q-106.339 23.066-105.983 23.066Q-105.761 23.066-105.577 22.982Q-105.392 22.898-105.257 22.743Q-105.122 22.587-105.064 22.379Q-105.050 22.324-104.996 22.324L-104.883 22.324Q-104.852 22.324-104.830 22.348Q-104.808 22.372-104.808 22.406L-104.808 22.427Q-104.893 22.714-105.081 22.912Q-105.269 23.110-105.534 23.213Q-105.799 23.315-106.093 23.315Q-106.523 23.315-106.911 23.109Q-107.299 22.902-107.533 22.539Q-107.768 22.177-107.768 21.736M-104.162 22.519Q-104.162 22.187-103.938 21.960Q-103.714 21.733-103.370 21.605Q-103.027 21.476-102.654 21.424Q-102.282 21.371-101.978 21.371L-101.978 21.118Q-101.978 20.913-102.085 20.733Q-102.193 20.554-102.374 20.451Q-102.555 20.349-102.764 20.349Q-103.170 20.349-103.406 20.441Q-103.317 20.478-103.271 20.562Q-103.225 20.646-103.225 20.748Q-103.225 20.844-103.271 20.923Q-103.317 21.001-103.398 21.046Q-103.478 21.090-103.567 21.090Q-103.717 21.090-103.818 20.993Q-103.919 20.895-103.919 20.748Q-103.919 20.126-102.764 20.126Q-102.552 20.126-102.302 20.190Q-102.053 20.253-101.851 20.372Q-101.649 20.492-101.523 20.677Q-101.396 20.861-101.396 21.104L-101.396 22.680Q-101.396 22.796-101.335 22.892Q-101.273 22.987-101.161 22.987Q-101.051 22.987-100.986 22.893Q-100.921 22.799-100.921 22.680L-100.921 22.232L-100.655 22.232L-100.655 22.680Q-100.655 22.950-100.882 23.115Q-101.109 23.281-101.390 23.281Q-101.598 23.281-101.735 23.127Q-101.872 22.974-101.895 22.758Q-102.042 23.025-102.324 23.170Q-102.606 23.315-102.931 23.315Q-103.208 23.315-103.492 23.240Q-103.775 23.165-103.968 22.986Q-104.162 22.806-104.162 22.519M-103.546 22.519Q-103.546 22.693-103.446 22.823Q-103.345 22.953-103.189 23.023Q-103.034 23.093-102.870 23.093Q-102.651 23.093-102.442 22.996Q-102.234 22.898-102.106 22.717Q-101.978 22.536-101.978 22.310L-101.978 21.582Q-102.302 21.582-102.668 21.673Q-103.034 21.764-103.290 21.976Q-103.546 22.187-103.546 22.519M-99.711 22.406L-99.711 20.509L-100.351 20.509L-100.351 20.287Q-100.033 20.287-99.816 20.077Q-99.599 19.867-99.498 19.557Q-99.397 19.248-99.397 18.940L-99.130 18.940L-99.130 20.229L-98.054 20.229L-98.054 20.509L-99.130 20.509L-99.130 22.393Q-99.130 22.669-99.026 22.868Q-98.922 23.066-98.662 23.066Q-98.505 23.066-98.399 22.962Q-98.293 22.857-98.243 22.704Q-98.194 22.550-98.194 22.393L-98.194 21.979L-97.927 21.979L-97.927 22.406Q-97.927 22.632-98.026 22.842Q-98.125 23.052-98.310 23.184Q-98.495 23.315-98.724 23.315Q-99.161 23.315-99.436 23.078Q-99.711 22.840-99.711 22.406\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M239.04 13.088V-6.674\"\u002F>\u003Cpath stroke=\"none\" d=\"m239.04-9.274-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M239.04-29.59v-19.763\"\u002F>\u003Cpath stroke=\"none\" d=\"m239.04-51.953-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M281.72-9.474h28.453v-19.917H281.72Z\"\u002F>\u003Cg transform=\"translate(398.265 -40.249)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.398-107.627 21.107Q-107.487 20.817-107.243 20.603Q-106.999 20.390-106.694 20.275Q-106.390 20.161-106.065 20.161Q-105.795 20.161-105.532 20.260Q-105.269 20.359-105.078 20.537L-105.078 19.139Q-105.078 18.869-105.185 18.807Q-105.293 18.746-105.604 18.746L-105.604 18.465L-104.527 18.390L-104.527 22.574Q-104.527 22.762-104.473 22.845Q-104.418 22.929-104.317 22.948Q-104.216 22.967-104.001 22.967L-104.001 23.247L-105.108 23.315L-105.108 22.898Q-105.525 23.315-106.151 23.315Q-106.582 23.315-106.954 23.103Q-107.327 22.892-107.547 22.531Q-107.768 22.170-107.768 21.736M-106.093 23.093Q-105.884 23.093-105.698 23.021Q-105.512 22.950-105.358 22.813Q-105.204 22.676-105.108 22.498L-105.108 20.889Q-105.194 20.742-105.339 20.622Q-105.484 20.502-105.654 20.443Q-105.823 20.383-106.004 20.383Q-106.564 20.383-106.833 20.772Q-107.101 21.162-107.101 21.743Q-107.101 22.314-106.867 22.704Q-106.633 23.093-106.093 23.093M-103.393 21.712Q-103.393 21.391-103.268 21.102Q-103.143 20.813-102.917 20.590Q-102.692 20.366-102.396 20.246Q-102.101 20.126-101.783 20.126Q-101.455 20.126-101.193 20.226Q-100.932 20.325-100.756 20.507Q-100.580 20.690-100.486 20.948Q-100.392 21.206-100.392 21.538Q-100.392 21.630-100.474 21.651L-102.729 21.651L-102.729 21.712Q-102.729 22.300-102.446 22.683Q-102.162 23.066-101.595 23.066Q-101.273 23.066-101.005 22.873Q-100.737 22.680-100.648 22.365Q-100.641 22.324-100.566 22.310L-100.474 22.310Q-100.392 22.334-100.392 22.406Q-100.392 22.413-100.398 22.440Q-100.511 22.837-100.882 23.076Q-101.253 23.315-101.677 23.315Q-102.114 23.315-102.514 23.107Q-102.914 22.898-103.153 22.531Q-103.393 22.164-103.393 21.712M-102.723 21.442L-100.908 21.442Q-100.908 21.165-101.005 20.913Q-101.103 20.660-101.301 20.504Q-101.499 20.349-101.783 20.349Q-102.060 20.349-102.273 20.507Q-102.487 20.666-102.605 20.921Q-102.723 21.176-102.723 21.442M-99.804 21.736Q-99.804 21.408-99.669 21.107Q-99.534 20.807-99.298 20.586Q-99.062 20.366-98.758 20.246Q-98.454 20.126-98.129 20.126Q-97.623 20.126-97.274 20.229Q-96.926 20.331-96.926 20.707Q-96.926 20.854-97.023 20.955Q-97.121 21.056-97.268 21.056Q-97.421 21.056-97.520 20.957Q-97.620 20.858-97.620 20.707Q-97.620 20.519-97.479 20.427Q-97.681 20.376-98.122 20.376Q-98.478 20.376-98.707 20.572Q-98.936 20.769-99.036 21.078Q-99.137 21.388-99.137 21.736Q-99.137 22.085-99.011 22.391Q-98.884 22.697-98.630 22.881Q-98.375 23.066-98.019 23.066Q-97.797 23.066-97.613 22.982Q-97.428 22.898-97.293 22.743Q-97.158 22.587-97.100 22.379Q-97.086 22.324-97.032 22.324L-96.919 22.324Q-96.888 22.324-96.866 22.348Q-96.844 22.372-96.844 22.406L-96.844 22.427Q-96.929 22.714-97.117 22.912Q-97.305 23.110-97.570 23.213Q-97.835 23.315-98.129 23.315Q-98.560 23.315-98.947 23.109Q-99.335 22.902-99.570 22.539Q-99.804 22.177-99.804 21.736\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M281.72 33.205h28.453V13.288H281.72Z\"\u002F>\u003Cg transform=\"translate(398.696 2.153)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.408-107.633 21.107Q-107.498 20.807-107.262 20.586Q-107.026 20.366-106.722 20.246Q-106.417 20.126-106.093 20.126Q-105.587 20.126-105.238 20.229Q-104.890 20.331-104.890 20.707Q-104.890 20.854-104.987 20.955Q-105.084 21.056-105.231 21.056Q-105.385 21.056-105.484 20.957Q-105.583 20.858-105.583 20.707Q-105.583 20.519-105.443 20.427Q-105.645 20.376-106.086 20.376Q-106.441 20.376-106.670 20.572Q-106.899 20.769-107 21.078Q-107.101 21.388-107.101 21.736Q-107.101 22.085-106.975 22.391Q-106.848 22.697-106.593 22.881Q-106.339 23.066-105.983 23.066Q-105.761 23.066-105.577 22.982Q-105.392 22.898-105.257 22.743Q-105.122 22.587-105.064 22.379Q-105.050 22.324-104.996 22.324L-104.883 22.324Q-104.852 22.324-104.830 22.348Q-104.808 22.372-104.808 22.406L-104.808 22.427Q-104.893 22.714-105.081 22.912Q-105.269 23.110-105.534 23.213Q-105.799 23.315-106.093 23.315Q-106.523 23.315-106.911 23.109Q-107.299 22.902-107.533 22.539Q-107.768 22.177-107.768 21.736M-104.162 22.519Q-104.162 22.187-103.938 21.960Q-103.714 21.733-103.370 21.605Q-103.027 21.476-102.654 21.424Q-102.282 21.371-101.978 21.371L-101.978 21.118Q-101.978 20.913-102.085 20.733Q-102.193 20.554-102.374 20.451Q-102.555 20.349-102.764 20.349Q-103.170 20.349-103.406 20.441Q-103.317 20.478-103.271 20.562Q-103.225 20.646-103.225 20.748Q-103.225 20.844-103.271 20.923Q-103.317 21.001-103.398 21.046Q-103.478 21.090-103.567 21.090Q-103.717 21.090-103.818 20.993Q-103.919 20.895-103.919 20.748Q-103.919 20.126-102.764 20.126Q-102.552 20.126-102.302 20.190Q-102.053 20.253-101.851 20.372Q-101.649 20.492-101.523 20.677Q-101.396 20.861-101.396 21.104L-101.396 22.680Q-101.396 22.796-101.335 22.892Q-101.273 22.987-101.161 22.987Q-101.051 22.987-100.986 22.893Q-100.921 22.799-100.921 22.680L-100.921 22.232L-100.655 22.232L-100.655 22.680Q-100.655 22.950-100.882 23.115Q-101.109 23.281-101.390 23.281Q-101.598 23.281-101.735 23.127Q-101.872 22.974-101.895 22.758Q-102.042 23.025-102.324 23.170Q-102.606 23.315-102.931 23.315Q-103.208 23.315-103.492 23.240Q-103.775 23.165-103.968 22.986Q-104.162 22.806-104.162 22.519M-103.546 22.519Q-103.546 22.693-103.446 22.823Q-103.345 22.953-103.189 23.023Q-103.034 23.093-102.870 23.093Q-102.651 23.093-102.442 22.996Q-102.234 22.898-102.106 22.717Q-101.978 22.536-101.978 22.310L-101.978 21.582Q-102.302 21.582-102.668 21.673Q-103.034 21.764-103.290 21.976Q-103.546 22.187-103.546 22.519M-99.711 22.406L-99.711 20.509L-100.351 20.509L-100.351 20.287Q-100.033 20.287-99.816 20.077Q-99.599 19.867-99.498 19.557Q-99.397 19.248-99.397 18.940L-99.130 18.940L-99.130 20.229L-98.054 20.229L-98.054 20.509L-99.130 20.509L-99.130 22.393Q-99.130 22.669-99.026 22.868Q-98.922 23.066-98.662 23.066Q-98.505 23.066-98.399 22.962Q-98.293 22.857-98.243 22.704Q-98.194 22.550-98.194 22.393L-98.194 21.979L-97.927 21.979L-97.927 22.406Q-97.927 22.632-98.026 22.842Q-98.125 23.052-98.310 23.184Q-98.495 23.315-98.724 23.315Q-99.161 23.315-99.436 23.078Q-99.711 22.840-99.711 22.406\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cpath fill=\"none\" d=\"M281.72-52.153h28.453V-72.07H281.72Z\"\u002F>\u003Cg transform=\"translate(393.965 -83.608)\">\u003Cpath d=\"M-107.768 23.240L-107.768 22.177Q-107.768 22.153-107.740 22.126Q-107.713 22.099-107.689 22.099L-107.580 22.099Q-107.515 22.099-107.501 22.157Q-107.405 22.591-107.159 22.842Q-106.913 23.093-106.499 23.093Q-106.158 23.093-105.905 22.960Q-105.652 22.827-105.652 22.519Q-105.652 22.362-105.746 22.247Q-105.840 22.133-105.978 22.064Q-106.117 21.996-106.284 21.958L-106.865 21.859Q-107.221 21.791-107.494 21.570Q-107.768 21.350-107.768 21.008Q-107.768 20.759-107.656 20.584Q-107.545 20.410-107.359 20.311Q-107.173 20.212-106.957 20.169Q-106.742 20.126-106.499 20.126Q-106.086 20.126-105.806 20.308L-105.590 20.133Q-105.580 20.130-105.573 20.128Q-105.566 20.126-105.556 20.126L-105.505 20.126Q-105.478 20.126-105.454 20.150Q-105.430 20.174-105.430 20.202L-105.430 21.049Q-105.430 21.070-105.454 21.097Q-105.478 21.124-105.505 21.124L-105.618 21.124Q-105.645 21.124-105.671 21.099Q-105.696 21.073-105.696 21.049Q-105.696 20.813-105.802 20.649Q-105.908 20.485-106.091 20.403Q-106.274 20.321-106.506 20.321Q-106.834 20.321-107.091 20.424Q-107.347 20.526-107.347 20.803Q-107.347 20.998-107.164 21.107Q-106.981 21.217-106.752 21.258L-106.178 21.364Q-105.932 21.412-105.718 21.540Q-105.505 21.668-105.368 21.871Q-105.231 22.075-105.231 22.324Q-105.231 22.837-105.597 23.076Q-105.963 23.315-106.499 23.315Q-106.995 23.315-107.327 23.021L-107.593 23.295Q-107.614 23.315-107.641 23.315L-107.689 23.315Q-107.713 23.315-107.740 23.288Q-107.768 23.261-107.768 23.240M-102.935 23.247L-104.538 23.247L-104.538 22.967Q-104.312 22.967-104.163 22.933Q-104.015 22.898-104.015 22.758L-104.015 19.139Q-104.015 18.869-104.122 18.807Q-104.230 18.746-104.538 18.746L-104.538 18.465L-103.461 18.390L-103.461 22.758Q-103.461 22.895-103.311 22.931Q-103.160 22.967-102.935 22.967L-102.935 23.247M-102.381 21.712Q-102.381 21.391-102.256 21.102Q-102.131 20.813-101.906 20.590Q-101.680 20.366-101.384 20.246Q-101.089 20.126-100.771 20.126Q-100.443 20.126-100.181 20.226Q-99.920 20.325-99.744 20.507Q-99.568 20.690-99.474 20.948Q-99.380 21.206-99.380 21.538Q-99.380 21.630-99.462 21.651L-101.718 21.651L-101.718 21.712Q-101.718 22.300-101.434 22.683Q-101.150 23.066-100.583 23.066Q-100.262 23.066-99.993 22.873Q-99.725 22.680-99.636 22.365Q-99.629 22.324-99.554 22.310L-99.462 22.310Q-99.380 22.334-99.380 22.406Q-99.380 22.413-99.387 22.440Q-99.499 22.837-99.870 23.076Q-100.241 23.315-100.665 23.315Q-101.103 23.315-101.502 23.107Q-101.902 22.898-102.142 22.531Q-102.381 22.164-102.381 21.712M-101.711 21.442L-99.896 21.442Q-99.896 21.165-99.993 20.913Q-100.091 20.660-100.289 20.504Q-100.487 20.349-100.771 20.349Q-101.048 20.349-101.261 20.507Q-101.475 20.666-101.593 20.921Q-101.711 21.176-101.711 21.442M-98.833 21.712Q-98.833 21.391-98.708 21.102Q-98.583 20.813-98.358 20.590Q-98.132 20.366-97.837 20.246Q-97.541 20.126-97.223 20.126Q-96.895 20.126-96.634 20.226Q-96.372 20.325-96.196 20.507Q-96.020 20.690-95.926 20.948Q-95.832 21.206-95.832 21.538Q-95.832 21.630-95.914 21.651L-98.170 21.651L-98.170 21.712Q-98.170 22.300-97.886 22.683Q-97.603 23.066-97.035 23.066Q-96.714 23.066-96.446 22.873Q-96.177 22.680-96.088 22.365Q-96.082 22.324-96.006 22.310L-95.914 22.310Q-95.832 22.334-95.832 22.406Q-95.832 22.413-95.839 22.440Q-95.952 22.837-96.322 23.076Q-96.693 23.315-97.117 23.315Q-97.555 23.315-97.955 23.107Q-98.354 22.898-98.594 22.531Q-98.833 22.164-98.833 21.712M-98.163 21.442L-96.348 21.442Q-96.348 21.165-96.446 20.913Q-96.543 20.660-96.741 20.504Q-96.939 20.349-97.223 20.349Q-97.500 20.349-97.714 20.507Q-97.927 20.666-98.045 20.921Q-98.163 21.176-98.163 21.442M-93.600 24.604L-95.230 24.604L-95.230 24.324Q-95.001 24.324-94.853 24.289Q-94.704 24.255-94.704 24.115L-94.704 20.769Q-94.704 20.598-94.841 20.557Q-94.978 20.516-95.230 20.516L-95.230 20.236L-94.150 20.161L-94.150 20.567Q-93.928 20.366-93.641 20.263Q-93.354 20.161-93.046 20.161Q-92.619 20.161-92.255 20.374Q-91.891 20.588-91.677 20.952Q-91.464 21.316-91.464 21.736Q-91.464 22.181-91.703 22.545Q-91.942 22.909-92.335 23.112Q-92.728 23.315-93.173 23.315Q-93.439 23.315-93.687 23.215Q-93.935 23.114-94.123 22.933L-94.123 24.115Q-94.123 24.252-93.974 24.288Q-93.826 24.324-93.600 24.324L-93.600 24.604M-94.123 20.916L-94.123 22.526Q-93.990 22.779-93.747 22.936Q-93.504 23.093-93.228 23.093Q-92.899 23.093-92.646 22.892Q-92.394 22.690-92.260 22.372Q-92.127 22.054-92.127 21.736Q-92.127 21.507-92.192 21.278Q-92.257 21.049-92.385 20.851Q-92.513 20.653-92.708 20.533Q-92.903 20.414-93.135 20.414Q-93.429 20.414-93.697 20.543Q-93.966 20.673-94.123 20.916M-90.828 23.240L-90.828 22.177Q-90.828 22.153-90.801 22.126Q-90.773 22.099-90.749 22.099L-90.640 22.099Q-90.575 22.099-90.561 22.157Q-90.466 22.591-90.220 22.842Q-89.974 23.093-89.560 23.093Q-89.218 23.093-88.965 22.960Q-88.712 22.827-88.712 22.519Q-88.712 22.362-88.806 22.247Q-88.900 22.133-89.039 22.064Q-89.177 21.996-89.345 21.958L-89.926 21.859Q-90.281 21.791-90.555 21.570Q-90.828 21.350-90.828 21.008Q-90.828 20.759-90.717 20.584Q-90.606 20.410-90.420 20.311Q-90.233 20.212-90.018 20.169Q-89.803 20.126-89.560 20.126Q-89.146 20.126-88.866 20.308L-88.651 20.133Q-88.641 20.130-88.634 20.128Q-88.627 20.126-88.617 20.126L-88.565 20.126Q-88.538 20.126-88.514 20.150Q-88.490 20.174-88.490 20.202L-88.490 21.049Q-88.490 21.070-88.514 21.097Q-88.538 21.124-88.565 21.124L-88.678 21.124Q-88.706 21.124-88.731 21.099Q-88.757 21.073-88.757 21.049Q-88.757 20.813-88.863 20.649Q-88.969 20.485-89.152 20.403Q-89.334 20.321-89.567 20.321Q-89.895 20.321-90.151 20.424Q-90.408 20.526-90.408 20.803Q-90.408 20.998-90.225 21.107Q-90.042 21.217-89.813 21.258L-89.239 21.364Q-88.993 21.412-88.779 21.540Q-88.565 21.668-88.429 21.871Q-88.292 22.075-88.292 22.324Q-88.292 22.837-88.658 23.076Q-89.023 23.315-89.560 23.315Q-90.056 23.315-90.387 23.021L-90.654 23.295Q-90.674 23.315-90.702 23.315L-90.749 23.315Q-90.773 23.315-90.801 23.288Q-90.828 23.261-90.828 23.240\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M295.946 13.088V-6.674\"\u002F>\u003Cpath stroke=\"none\" d=\"m295.946-9.274-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M295.946-29.59v-19.763\"\u002F>\u003Cpath stroke=\"none\" d=\"m295.946-51.953-2.08 4.16 2.08-1.56 2.08 1.56\"\u002F>\u003C\u002Fg>\u003Cg fill=\"var(--tk-accent)\" stroke=\"var(--tk-accent)\" style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M141.279-19.432h23.83\"\u002F>\u003Cpath stroke=\"none\" d=\"m167.71-19.432-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M196.562-19.432h25.453\"\u002F>\u003Cpath stroke=\"none\" d=\"m224.615-19.432-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg style=\"stroke-width:.8\">\u003Cpath fill=\"none\" d=\"M253.467-19.432h25.453\"\u002F>\u003Cpath stroke=\"none\" d=\"m281.52-19.432-4.16-2.08 1.56 2.08-1.56 2.08\"\u002F>\u003C\u002Fg>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(85.022 31.545)\">\u003Cpath d=\"M-107.809 21.712Q-107.809 21.391-107.684 21.102Q-107.559 20.813-107.333 20.590Q-107.108 20.366-106.812 20.246Q-106.517 20.126-106.199 20.126Q-105.871 20.126-105.609 20.226Q-105.348 20.325-105.172 20.507Q-104.996 20.690-104.902 20.948Q-104.808 21.206-104.808 21.538Q-104.808 21.630-104.890 21.651L-107.145 21.651L-107.145 21.712Q-107.145 22.300-106.862 22.683Q-106.578 23.066-106.011 23.066Q-105.689 23.066-105.421 22.873Q-105.153 22.680-105.064 22.365Q-105.057 22.324-104.982 22.310L-104.890 22.310Q-104.808 22.334-104.808 22.406Q-104.808 22.413-104.814 22.440Q-104.927 22.837-105.298 23.076Q-105.669 23.315-106.093 23.315Q-106.530 23.315-106.930 23.107Q-107.330 22.898-107.569 22.531Q-107.809 22.164-107.809 21.712M-107.139 21.442L-105.324 21.442Q-105.324 21.165-105.421 20.913Q-105.519 20.660-105.717 20.504Q-105.915 20.349-106.199 20.349Q-106.476 20.349-106.689 20.507Q-106.903 20.666-107.021 20.921Q-107.139 21.176-107.139 21.442M-102.538 23.247L-104.172 23.247L-104.172 22.967Q-103.943 22.967-103.794 22.933Q-103.645 22.898-103.645 22.758L-103.645 20.909Q-103.645 20.639-103.753 20.578Q-103.861 20.516-104.172 20.516L-104.172 20.236L-103.112 20.161L-103.112 20.810Q-102.941 20.502-102.637 20.331Q-102.333 20.161-101.988 20.161Q-101.482 20.161-101.198 20.384Q-100.915 20.608-100.915 21.104L-100.915 22.758Q-100.915 22.895-100.766 22.931Q-100.617 22.967-100.392 22.967L-100.392 23.247L-102.022 23.247L-102.022 22.967Q-101.793 22.967-101.644 22.933Q-101.496 22.898-101.496 22.758L-101.496 21.118Q-101.496 20.783-101.615 20.583Q-101.735 20.383-102.049 20.383Q-102.319 20.383-102.553 20.519Q-102.788 20.656-102.926 20.890Q-103.064 21.124-103.064 21.398L-103.064 22.758Q-103.064 22.895-102.914 22.931Q-102.764 22.967-102.538 22.967L-102.538 23.247M-99.804 21.736Q-99.804 21.408-99.669 21.107Q-99.534 20.807-99.298 20.586Q-99.062 20.366-98.758 20.246Q-98.454 20.126-98.129 20.126Q-97.623 20.126-97.274 20.229Q-96.926 20.331-96.926 20.707Q-96.926 20.854-97.023 20.955Q-97.121 21.056-97.268 21.056Q-97.421 21.056-97.520 20.957Q-97.620 20.858-97.620 20.707Q-97.620 20.519-97.479 20.427Q-97.681 20.376-98.122 20.376Q-98.478 20.376-98.707 20.572Q-98.936 20.769-99.036 21.078Q-99.137 21.388-99.137 21.736Q-99.137 22.085-99.011 22.391Q-98.884 22.697-98.630 22.881Q-98.375 23.066-98.019 23.066Q-97.797 23.066-97.613 22.982Q-97.428 22.898-97.293 22.743Q-97.158 22.587-97.100 22.379Q-97.086 22.324-97.032 22.324L-96.919 22.324Q-96.888 22.324-96.866 22.348Q-96.844 22.372-96.844 22.406L-96.844 22.427Q-96.929 22.714-97.117 22.912Q-97.305 23.110-97.570 23.213Q-97.835 23.315-98.129 23.315Q-98.560 23.315-98.947 23.109Q-99.335 22.902-99.570 22.539Q-99.804 22.177-99.804 21.736M-96.297 21.764Q-96.297 21.422-96.162 21.123Q-96.027 20.824-95.788 20.600Q-95.548 20.376-95.230 20.251Q-94.913 20.126-94.581 20.126Q-94.137 20.126-93.737 20.342Q-93.337 20.557-93.103 20.935Q-92.869 21.312-92.869 21.764Q-92.869 22.105-93.010 22.389Q-93.152 22.673-93.397 22.880Q-93.641 23.086-93.950 23.201Q-94.260 23.315-94.581 23.315Q-95.012 23.315-95.413 23.114Q-95.815 22.912-96.056 22.560Q-96.297 22.208-96.297 21.764M-94.581 23.066Q-93.979 23.066-93.756 22.688Q-93.532 22.310-93.532 21.678Q-93.532 21.066-93.766 20.707Q-94 20.349-94.581 20.349Q-95.634 20.349-95.634 21.678Q-95.634 22.310-95.408 22.688Q-95.183 23.066-94.581 23.066\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(85.022 31.545)\">\u003Cpath d=\"M-92.039 21.736Q-92.039 21.398-91.898 21.107Q-91.758 20.817-91.514 20.603Q-91.270 20.390-90.965 20.275Q-90.661 20.161-90.336 20.161Q-90.066 20.161-89.803 20.260Q-89.540 20.359-89.349 20.537L-89.349 19.139Q-89.349 18.869-89.456 18.807Q-89.564 18.746-89.875 18.746L-89.875 18.465L-88.798 18.390L-88.798 22.574Q-88.798 22.762-88.744 22.845Q-88.689 22.929-88.588 22.948Q-88.487 22.967-88.272 22.967L-88.272 23.247L-89.379 23.315L-89.379 22.898Q-89.796 23.315-90.422 23.315Q-90.853 23.315-91.225 23.103Q-91.598 22.892-91.818 22.531Q-92.039 22.170-92.039 21.736M-90.364 23.093Q-90.155 23.093-89.969 23.021Q-89.783 22.950-89.629 22.813Q-89.475 22.676-89.379 22.498L-89.379 20.889Q-89.465 20.742-89.610 20.622Q-89.755 20.502-89.925 20.443Q-90.094 20.383-90.275 20.383Q-90.835 20.383-91.104 20.772Q-91.372 21.162-91.372 21.743Q-91.372 22.314-91.138 22.704Q-90.904 23.093-90.364 23.093M-87.664 21.712Q-87.664 21.391-87.539 21.102Q-87.414 20.813-87.188 20.590Q-86.963 20.366-86.667 20.246Q-86.372 20.126-86.054 20.126Q-85.726 20.126-85.464 20.226Q-85.203 20.325-85.027 20.507Q-84.851 20.690-84.757 20.948Q-84.663 21.206-84.663 21.538Q-84.663 21.630-84.745 21.651L-87 21.651L-87 21.712Q-87 22.300-86.717 22.683Q-86.433 23.066-85.866 23.066Q-85.544 23.066-85.276 22.873Q-85.008 22.680-84.919 22.365Q-84.912 22.324-84.837 22.310L-84.745 22.310Q-84.663 22.334-84.663 22.406Q-84.663 22.413-84.669 22.440Q-84.782 22.837-85.153 23.076Q-85.524 23.315-85.948 23.315Q-86.385 23.315-86.785 23.107Q-87.185 22.898-87.424 22.531Q-87.664 22.164-87.664 21.712M-86.994 21.442L-85.179 21.442Q-85.179 21.165-85.276 20.913Q-85.374 20.660-85.572 20.504Q-85.770 20.349-86.054 20.349Q-86.331 20.349-86.544 20.507Q-86.758 20.666-86.876 20.921Q-86.994 21.176-86.994 21.442M-82.325 23.247L-84.061 23.247L-84.061 22.967Q-83.832 22.967-83.683 22.933Q-83.535 22.898-83.535 22.758L-83.535 20.909Q-83.535 20.639-83.642 20.578Q-83.750 20.516-84.061 20.516L-84.061 20.236L-83.032 20.161L-83.032 20.868Q-82.902 20.560-82.660 20.361Q-82.417 20.161-82.099 20.161Q-81.880 20.161-81.709 20.285Q-81.539 20.410-81.539 20.622Q-81.539 20.759-81.638 20.858Q-81.737 20.957-81.870 20.957Q-82.007 20.957-82.106 20.858Q-82.205 20.759-82.205 20.622Q-82.205 20.482-82.106 20.383Q-82.396 20.383-82.596 20.579Q-82.796 20.776-82.889 21.070Q-82.981 21.364-82.981 21.644L-82.981 22.758Q-82.981 22.967-82.325 22.967\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(85.022 31.545)\">\u003Cpath d=\"M-76.117 24.997Q-76.667 24.597-77.038 24.042Q-77.409 23.486-77.590 22.840Q-77.771 22.194-77.771 21.497Q-77.771 20.984-77.671 20.489Q-77.570 19.993-77.365 19.542Q-77.160 19.091-76.847 18.699Q-76.534 18.308-76.117 18.004Q-76.107 18-76.100 17.999Q-76.093 17.997-76.083 17.997L-76.015 17.997Q-75.980 17.997-75.958 18.021Q-75.936 18.045-75.936 18.082Q-75.936 18.127-75.963 18.144Q-76.312 18.445-76.565 18.829Q-76.818 19.214-76.970 19.655Q-77.122 20.096-77.194 20.552Q-77.266 21.008-77.266 21.497Q-77.266 22.498-76.956 23.385Q-76.647 24.272-75.963 24.857Q-75.936 24.874-75.936 24.918Q-75.936 24.956-75.958 24.980Q-75.980 25.004-76.015 25.004L-76.083 25.004Q-76.090 25-76.098 24.999Q-76.107 24.997-76.117 24.997M-75.126 23.240L-75.126 22.177Q-75.126 22.153-75.099 22.126Q-75.071 22.099-75.047 22.099L-74.938 22.099Q-74.873 22.099-74.859 22.157Q-74.764 22.591-74.518 22.842Q-74.271 23.093-73.858 23.093Q-73.516 23.093-73.263 22.960Q-73.010 22.827-73.010 22.519Q-73.010 22.362-73.104 22.247Q-73.198 22.133-73.337 22.064Q-73.475 21.996-73.643 21.958L-74.224 21.859Q-74.579 21.791-74.853 21.570Q-75.126 21.350-75.126 21.008Q-75.126 20.759-75.015 20.584Q-74.904 20.410-74.717 20.311Q-74.531 20.212-74.316 20.169Q-74.101 20.126-73.858 20.126Q-73.444 20.126-73.164 20.308L-72.949 20.133Q-72.938 20.130-72.932 20.128Q-72.925 20.126-72.915 20.126L-72.863 20.126Q-72.836 20.126-72.812 20.150Q-72.788 20.174-72.788 20.202L-72.788 21.049Q-72.788 21.070-72.812 21.097Q-72.836 21.124-72.863 21.124L-72.976 21.124Q-73.003 21.124-73.029 21.099Q-73.055 21.073-73.055 21.049Q-73.055 20.813-73.161 20.649Q-73.267 20.485-73.449 20.403Q-73.632 20.321-73.865 20.321Q-74.193 20.321-74.449 20.424Q-74.706 20.526-74.706 20.803Q-74.706 20.998-74.523 21.107Q-74.340 21.217-74.111 21.258L-73.537 21.364Q-73.290 21.412-73.077 21.540Q-72.863 21.668-72.727 21.871Q-72.590 22.075-72.590 22.324Q-72.590 22.837-72.956 23.076Q-73.321 23.315-73.858 23.315Q-74.353 23.315-74.685 23.021L-74.952 23.295Q-74.972 23.315-74.999 23.315L-75.047 23.315Q-75.071 23.315-75.099 23.288Q-75.126 23.261-75.126 23.240M-72.002 21.764Q-72.002 21.422-71.867 21.123Q-71.732 20.824-71.493 20.600Q-71.253 20.376-70.936 20.251Q-70.618 20.126-70.286 20.126Q-69.842 20.126-69.442 20.342Q-69.042 20.557-68.808 20.935Q-68.574 21.312-68.574 21.764Q-68.574 22.105-68.716 22.389Q-68.857 22.673-69.102 22.880Q-69.346 23.086-69.655 23.201Q-69.965 23.315-70.286 23.315Q-70.717 23.315-71.118 23.114Q-71.520 22.912-71.761 22.560Q-72.002 22.208-72.002 21.764M-70.286 23.066Q-69.685 23.066-69.461 22.688Q-69.237 22.310-69.237 21.678Q-69.237 21.066-69.471 20.707Q-69.705 20.349-70.286 20.349Q-71.339 20.349-71.339 21.678Q-71.339 22.310-71.113 22.688Q-70.888 23.066-70.286 23.066M-67.405 22.413L-67.405 20.909Q-67.405 20.639-67.512 20.578Q-67.620 20.516-67.931 20.516L-67.931 20.236L-66.824 20.161L-66.824 22.393L-66.824 22.413Q-66.824 22.693-66.772 22.837Q-66.721 22.980-66.579 23.037Q-66.437 23.093-66.150 23.093Q-65.897 23.093-65.692 22.953Q-65.487 22.813-65.371 22.587Q-65.255 22.362-65.255 22.112L-65.255 20.909Q-65.255 20.639-65.363 20.578Q-65.470 20.516-65.781 20.516L-65.781 20.236L-64.674 20.161L-64.674 22.574Q-64.674 22.765-64.621 22.847Q-64.568 22.929-64.467 22.948Q-64.366 22.967-64.151 22.967L-64.151 23.247L-65.228 23.315L-65.228 22.751Q-65.337 22.933-65.482 23.056Q-65.627 23.179-65.814 23.247Q-66 23.315-66.202 23.315Q-67.405 23.315-67.405 22.413M-61.813 23.247L-63.549 23.247L-63.549 22.967Q-63.320 22.967-63.172 22.933Q-63.023 22.898-63.023 22.758L-63.023 20.909Q-63.023 20.639-63.131 20.578Q-63.238 20.516-63.549 20.516L-63.549 20.236L-62.520 20.161L-62.520 20.868Q-62.391 20.560-62.148 20.361Q-61.905 20.161-61.587 20.161Q-61.369 20.161-61.198 20.285Q-61.027 20.410-61.027 20.622Q-61.027 20.759-61.126 20.858Q-61.225 20.957-61.358 20.957Q-61.495 20.957-61.594 20.858Q-61.693 20.759-61.693 20.622Q-61.693 20.482-61.594 20.383Q-61.885 20.383-62.085 20.579Q-62.285 20.776-62.377 21.070Q-62.469 21.364-62.469 21.644L-62.469 22.758Q-62.469 22.967-61.813 22.967L-61.813 23.247M-60.442 21.736Q-60.442 21.408-60.307 21.107Q-60.172 20.807-59.936 20.586Q-59.701 20.366-59.396 20.246Q-59.092 20.126-58.768 20.126Q-58.262 20.126-57.913 20.229Q-57.564 20.331-57.564 20.707Q-57.564 20.854-57.662 20.955Q-57.759 21.056-57.906 21.056Q-58.060 21.056-58.159 20.957Q-58.258 20.858-58.258 20.707Q-58.258 20.519-58.118 20.427Q-58.320 20.376-58.761 20.376Q-59.116 20.376-59.345 20.572Q-59.574 20.769-59.675 21.078Q-59.776 21.388-59.776 21.736Q-59.776 22.085-59.649 22.391Q-59.523 22.697-59.268 22.881Q-59.014 23.066-58.658 23.066Q-58.436 23.066-58.251 22.982Q-58.067 22.898-57.932 22.743Q-57.797 22.587-57.739 22.379Q-57.725 22.324-57.670 22.324L-57.558 22.324Q-57.527 22.324-57.505 22.348Q-57.482 22.372-57.482 22.406L-57.482 22.427Q-57.568 22.714-57.756 22.912Q-57.944 23.110-58.209 23.213Q-58.474 23.315-58.768 23.315Q-59.198 23.315-59.586 23.109Q-59.974 22.902-60.208 22.539Q-60.442 22.177-60.442 21.736M-56.936 21.712Q-56.936 21.391-56.811 21.102Q-56.686 20.813-56.460 20.590Q-56.235 20.366-55.939 20.246Q-55.644 20.126-55.326 20.126Q-54.998 20.126-54.736 20.226Q-54.475 20.325-54.299 20.507Q-54.123 20.690-54.029 20.948Q-53.935 21.206-53.935 21.538Q-53.935 21.630-54.017 21.651L-56.272 21.651L-56.272 21.712Q-56.272 22.300-55.989 22.683Q-55.705 23.066-55.138 23.066Q-54.816 23.066-54.548 22.873Q-54.280 22.680-54.191 22.365Q-54.184 22.324-54.109 22.310L-54.017 22.310Q-53.935 22.334-53.935 22.406Q-53.935 22.413-53.941 22.440Q-54.054 22.837-54.425 23.076Q-54.796 23.315-55.220 23.315Q-55.657 23.315-56.057 23.107Q-56.457 22.898-56.696 22.531Q-56.936 22.164-56.936 21.712M-56.266 21.442L-54.451 21.442Q-54.451 21.165-54.548 20.913Q-54.645 20.660-54.844 20.504Q-55.042 20.349-55.326 20.349Q-55.603 20.349-55.816 20.507Q-56.030 20.666-56.148 20.921Q-56.266 21.176-56.266 21.442M-53.025 25.004L-53.094 25.004Q-53.128 25.004-53.150 24.978Q-53.172 24.953-53.172 24.918Q-53.172 24.874-53.142 24.857Q-52.786 24.553-52.537 24.163Q-52.287 23.773-52.135 23.341Q-51.983 22.909-51.913 22.440Q-51.843 21.972-51.843 21.497Q-51.843 21.018-51.913 20.552Q-51.983 20.085-52.137 19.650Q-52.290 19.214-52.542 18.826Q-52.793 18.438-53.142 18.144Q-53.172 18.127-53.172 18.082Q-53.172 18.048-53.150 18.023Q-53.128 17.997-53.094 17.997L-53.025 17.997Q-53.015 17.997-53.007 17.999Q-52.998 18-52.988 18.004Q-52.444 18.404-52.072 18.957Q-51.699 19.511-51.518 20.157Q-51.337 20.803-51.337 21.497Q-51.337 22.198-51.518 22.845Q-51.699 23.493-52.073 24.047Q-52.448 24.601-52.988 24.997Q-52.998 24.997-53.007 24.999Q-53.015 25-53.025 25.004\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003Cg stroke=\"none\" font-family=\"cmr7\" font-size=\"7\">\u003Cg transform=\"translate(318.786 31.545)\">\u003Cpath d=\"M-107.768 21.736Q-107.768 21.398-107.627 21.107Q-107.487 20.817-107.243 20.603Q-106.999 20.390-106.694 20.275Q-106.390 20.161-106.065 20.161Q-105.795 20.161-105.532 20.260Q-105.269 20.359-105.078 20.537L-105.078 19.139Q-105.078 18.869-105.185 18.807Q-105.293 18.746-105.604 18.746L-105.604 18.465L-104.527 18.390L-104.527 22.574Q-104.527 22.762-104.473 22.845Q-104.418 22.929-104.317 22.948Q-104.216 22.967-104.001 22.967L-104.001 23.247L-105.108 23.315L-105.108 22.898Q-105.525 23.315-106.151 23.315Q-106.582 23.315-106.954 23.103Q-107.327 22.892-107.547 22.531Q-107.768 22.170-107.768 21.736M-106.093 23.093Q-105.884 23.093-105.698 23.021Q-105.512 22.950-105.358 22.813Q-105.204 22.676-105.108 22.498L-105.108 20.889Q-105.194 20.742-105.339 20.622Q-105.484 20.502-105.654 20.443Q-105.823 20.383-106.004 20.383Q-106.564 20.383-106.833 20.772Q-107.101 21.162-107.101 21.743Q-107.101 22.314-106.867 22.704Q-106.633 23.093-106.093 23.093M-103.393 21.712Q-103.393 21.391-103.268 21.102Q-103.143 20.813-102.917 20.590Q-102.692 20.366-102.396 20.246Q-102.101 20.126-101.783 20.126Q-101.455 20.126-101.193 20.226Q-100.932 20.325-100.756 20.507Q-100.580 20.690-100.486 20.948Q-100.392 21.206-100.392 21.538Q-100.392 21.630-100.474 21.651L-102.729 21.651L-102.729 21.712Q-102.729 22.300-102.446 22.683Q-102.162 23.066-101.595 23.066Q-101.273 23.066-101.005 22.873Q-100.737 22.680-100.648 22.365Q-100.641 22.324-100.566 22.310L-100.474 22.310Q-100.392 22.334-100.392 22.406Q-100.392 22.413-100.398 22.440Q-100.511 22.837-100.882 23.076Q-101.253 23.315-101.677 23.315Q-102.114 23.315-102.514 23.107Q-102.914 22.898-103.153 22.531Q-103.393 22.164-103.393 21.712M-102.723 21.442L-100.908 21.442Q-100.908 21.165-101.005 20.913Q-101.103 20.660-101.301 20.504Q-101.499 20.349-101.783 20.349Q-102.060 20.349-102.273 20.507Q-102.487 20.666-102.605 20.921Q-102.723 21.176-102.723 21.442M-99.804 21.736Q-99.804 21.408-99.669 21.107Q-99.534 20.807-99.298 20.586Q-99.062 20.366-98.758 20.246Q-98.454 20.126-98.129 20.126Q-97.623 20.126-97.274 20.229Q-96.926 20.331-96.926 20.707Q-96.926 20.854-97.023 20.955Q-97.121 21.056-97.268 21.056Q-97.421 21.056-97.520 20.957Q-97.620 20.858-97.620 20.707Q-97.620 20.519-97.479 20.427Q-97.681 20.376-98.122 20.376Q-98.478 20.376-98.707 20.572Q-98.936 20.769-99.036 21.078Q-99.137 21.388-99.137 21.736Q-99.137 22.085-99.011 22.391Q-98.884 22.697-98.630 22.881Q-98.375 23.066-98.019 23.066Q-97.797 23.066-97.613 22.982Q-97.428 22.898-97.293 22.743Q-97.158 22.587-97.100 22.379Q-97.086 22.324-97.032 22.324L-96.919 22.324Q-96.888 22.324-96.866 22.348Q-96.844 22.372-96.844 22.406L-96.844 22.427Q-96.929 22.714-97.117 22.912Q-97.305 23.110-97.570 23.213Q-97.835 23.315-98.129 23.315Q-98.560 23.315-98.947 23.109Q-99.335 22.902-99.570 22.539Q-99.804 22.177-99.804 21.736M-96.297 21.764Q-96.297 21.422-96.162 21.123Q-96.027 20.824-95.788 20.600Q-95.548 20.376-95.230 20.251Q-94.913 20.126-94.581 20.126Q-94.137 20.126-93.737 20.342Q-93.337 20.557-93.103 20.935Q-92.869 21.312-92.869 21.764Q-92.869 22.105-93.010 22.389Q-93.152 22.673-93.397 22.880Q-93.641 23.086-93.950 23.201Q-94.260 23.315-94.581 23.315Q-95.012 23.315-95.413 23.114Q-95.815 22.912-96.056 22.560Q-96.297 22.208-96.297 21.764M-94.581 23.066Q-93.979 23.066-93.756 22.688Q-93.532 22.310-93.532 21.678Q-93.532 21.066-93.766 20.707Q-94 20.349-94.581 20.349Q-95.634 20.349-95.634 21.678Q-95.634 22.310-95.408 22.688Q-95.183 23.066-94.581 23.066\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(318.786 31.545)\">\u003Cpath d=\"M-92.039 21.736Q-92.039 21.398-91.898 21.107Q-91.758 20.817-91.514 20.603Q-91.270 20.390-90.965 20.275Q-90.661 20.161-90.336 20.161Q-90.066 20.161-89.803 20.260Q-89.540 20.359-89.349 20.537L-89.349 19.139Q-89.349 18.869-89.456 18.807Q-89.564 18.746-89.875 18.746L-89.875 18.465L-88.798 18.390L-88.798 22.574Q-88.798 22.762-88.744 22.845Q-88.689 22.929-88.588 22.948Q-88.487 22.967-88.272 22.967L-88.272 23.247L-89.379 23.315L-89.379 22.898Q-89.796 23.315-90.422 23.315Q-90.853 23.315-91.225 23.103Q-91.598 22.892-91.818 22.531Q-92.039 22.170-92.039 21.736M-90.364 23.093Q-90.155 23.093-89.969 23.021Q-89.783 22.950-89.629 22.813Q-89.475 22.676-89.379 22.498L-89.379 20.889Q-89.465 20.742-89.610 20.622Q-89.755 20.502-89.925 20.443Q-90.094 20.383-90.275 20.383Q-90.835 20.383-91.104 20.772Q-91.372 21.162-91.372 21.743Q-91.372 22.314-91.138 22.704Q-90.904 23.093-90.364 23.093M-87.664 21.712Q-87.664 21.391-87.539 21.102Q-87.414 20.813-87.188 20.590Q-86.963 20.366-86.667 20.246Q-86.372 20.126-86.054 20.126Q-85.726 20.126-85.464 20.226Q-85.203 20.325-85.027 20.507Q-84.851 20.690-84.757 20.948Q-84.663 21.206-84.663 21.538Q-84.663 21.630-84.745 21.651L-87 21.651L-87 21.712Q-87 22.300-86.717 22.683Q-86.433 23.066-85.866 23.066Q-85.544 23.066-85.276 22.873Q-85.008 22.680-84.919 22.365Q-84.912 22.324-84.837 22.310L-84.745 22.310Q-84.663 22.334-84.663 22.406Q-84.663 22.413-84.669 22.440Q-84.782 22.837-85.153 23.076Q-85.524 23.315-85.948 23.315Q-86.385 23.315-86.785 23.107Q-87.185 22.898-87.424 22.531Q-87.664 22.164-87.664 21.712M-86.994 21.442L-85.179 21.442Q-85.179 21.165-85.276 20.913Q-85.374 20.660-85.572 20.504Q-85.770 20.349-86.054 20.349Q-86.331 20.349-86.544 20.507Q-86.758 20.666-86.876 20.921Q-86.994 21.176-86.994 21.442M-82.325 23.247L-84.061 23.247L-84.061 22.967Q-83.832 22.967-83.683 22.933Q-83.535 22.898-83.535 22.758L-83.535 20.909Q-83.535 20.639-83.642 20.578Q-83.750 20.516-84.061 20.516L-84.061 20.236L-83.032 20.161L-83.032 20.868Q-82.902 20.560-82.660 20.361Q-82.417 20.161-82.099 20.161Q-81.880 20.161-81.709 20.285Q-81.539 20.410-81.539 20.622Q-81.539 20.759-81.638 20.858Q-81.737 20.957-81.870 20.957Q-82.007 20.957-82.106 20.858Q-82.205 20.759-82.205 20.622Q-82.205 20.482-82.106 20.383Q-82.396 20.383-82.596 20.579Q-82.796 20.776-82.889 21.070Q-82.981 21.364-82.981 21.644L-82.981 22.758Q-82.981 22.967-82.325 22.967\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003Cg transform=\"translate(318.786 31.545)\">\u003Cpath d=\"M-76.117 24.997Q-76.667 24.597-77.038 24.042Q-77.409 23.486-77.590 22.840Q-77.771 22.194-77.771 21.497Q-77.771 20.984-77.671 20.489Q-77.570 19.993-77.365 19.542Q-77.160 19.091-76.847 18.699Q-76.534 18.308-76.117 18.004Q-76.107 18-76.100 17.999Q-76.093 17.997-76.083 17.997L-76.015 17.997Q-75.980 17.997-75.958 18.021Q-75.936 18.045-75.936 18.082Q-75.936 18.127-75.963 18.144Q-76.312 18.445-76.565 18.829Q-76.818 19.214-76.970 19.655Q-77.122 20.096-77.194 20.552Q-77.266 21.008-77.266 21.497Q-77.266 22.498-76.956 23.385Q-76.647 24.272-75.963 24.857Q-75.936 24.874-75.936 24.918Q-75.936 24.956-75.958 24.980Q-75.980 25.004-76.015 25.004L-76.083 25.004Q-76.090 25-76.098 24.999Q-76.107 24.997-76.117 24.997M-74.600 22.406L-74.600 20.509L-75.239 20.509L-75.239 20.287Q-74.921 20.287-74.704 20.077Q-74.487 19.867-74.386 19.557Q-74.285 19.248-74.285 18.940L-74.019 18.940L-74.019 20.229L-72.942 20.229L-72.942 20.509L-74.019 20.509L-74.019 22.393Q-74.019 22.669-73.914 22.868Q-73.810 23.066-73.550 23.066Q-73.393 23.066-73.287 22.962Q-73.181 22.857-73.132 22.704Q-73.082 22.550-73.082 22.393L-73.082 21.979L-72.815 21.979L-72.815 22.406Q-72.815 22.632-72.915 22.842Q-73.014 23.052-73.198 23.184Q-73.383 23.315-73.612 23.315Q-74.049 23.315-74.324 23.078Q-74.600 22.840-74.600 22.406M-71.947 22.519Q-71.947 22.187-71.723 21.960Q-71.499 21.733-71.156 21.605Q-70.812 21.476-70.440 21.424Q-70.067 21.371-69.763 21.371L-69.763 21.118Q-69.763 20.913-69.871 20.733Q-69.978 20.554-70.160 20.451Q-70.341 20.349-70.549 20.349Q-70.956 20.349-71.192 20.441Q-71.103 20.478-71.057 20.562Q-71.011 20.646-71.011 20.748Q-71.011 20.844-71.057 20.923Q-71.103 21.001-71.183 21.046Q-71.264 21.090-71.353 21.090Q-71.503 21.090-71.604 20.993Q-71.705 20.895-71.705 20.748Q-71.705 20.126-70.549 20.126Q-70.337 20.126-70.088 20.190Q-69.838 20.253-69.637 20.372Q-69.435 20.492-69.309 20.677Q-69.182 20.861-69.182 21.104L-69.182 22.680Q-69.182 22.796-69.121 22.892Q-69.059 22.987-68.946 22.987Q-68.837 22.987-68.772 22.893Q-68.707 22.799-68.707 22.680L-68.707 22.232L-68.440 22.232L-68.440 22.680Q-68.440 22.950-68.668 23.115Q-68.895 23.281-69.175 23.281Q-69.384 23.281-69.520 23.127Q-69.657 22.974-69.681 22.758Q-69.828 23.025-70.110 23.170Q-70.392 23.315-70.717 23.315Q-70.994 23.315-71.277 23.240Q-71.561 23.165-71.754 22.986Q-71.947 22.806-71.947 22.519M-71.332 22.519Q-71.332 22.693-71.231 22.823Q-71.130 22.953-70.975 23.023Q-70.819 23.093-70.655 23.093Q-70.436 23.093-70.228 22.996Q-70.019 22.898-69.891 22.717Q-69.763 22.536-69.763 22.310L-69.763 21.582Q-70.088 21.582-70.454 21.673Q-70.819 21.764-71.076 21.976Q-71.332 22.187-71.332 22.519M-66.273 23.247L-68.010 23.247L-68.010 22.967Q-67.781 22.967-67.632 22.933Q-67.483 22.898-67.483 22.758L-67.483 20.909Q-67.483 20.639-67.591 20.578Q-67.699 20.516-68.010 20.516L-68.010 20.236L-66.981 20.161L-66.981 20.868Q-66.851 20.560-66.608 20.361Q-66.366 20.161-66.048 20.161Q-65.829 20.161-65.658 20.285Q-65.487 20.410-65.487 20.622Q-65.487 20.759-65.586 20.858Q-65.686 20.957-65.819 20.957Q-65.956 20.957-66.055 20.858Q-66.154 20.759-66.154 20.622Q-66.154 20.482-66.055 20.383Q-66.345 20.383-66.545 20.579Q-66.745 20.776-66.837 21.070Q-66.930 21.364-66.930 21.644L-66.930 22.758Q-66.930 22.967-66.273 22.967L-66.273 23.247M-64.944 23.780Q-64.944 23.534-64.747 23.350Q-64.551 23.165-64.294 23.086Q-64.431 22.974-64.503 22.813Q-64.575 22.652-64.575 22.471Q-64.575 22.150-64.363 21.904Q-64.698 21.606-64.698 21.196Q-64.698 20.735-64.308 20.448Q-63.918 20.161-63.440 20.161Q-62.968 20.161-62.633 20.407Q-62.459 20.253-62.249 20.171Q-62.039 20.089-61.810 20.089Q-61.645 20.089-61.524 20.196Q-61.403 20.304-61.403 20.468Q-61.403 20.564-61.475 20.636Q-61.546 20.707-61.639 20.707Q-61.738 20.707-61.808 20.634Q-61.878 20.560-61.878 20.461Q-61.878 20.407-61.864 20.376L-61.857 20.362Q-61.851 20.342-61.842 20.331Q-61.833 20.321-61.830 20.314Q-62.186 20.314-62.473 20.537Q-62.186 20.830-62.186 21.196Q-62.186 21.511-62.370 21.743Q-62.555 21.976-62.843 22.104Q-63.132 22.232-63.440 22.232Q-63.642 22.232-63.833 22.182Q-64.024 22.133-64.202 22.023Q-64.294 22.150-64.294 22.293Q-64.294 22.475-64.166 22.610Q-64.038 22.745-63.853 22.745L-63.221 22.745Q-62.773 22.745-62.404 22.816Q-62.035 22.888-61.775 23.117Q-61.516 23.346-61.516 23.780Q-61.516 24.101-61.811 24.303Q-62.107 24.505-62.510 24.594Q-62.914 24.683-63.228 24.683Q-63.546 24.683-63.949 24.594Q-64.353 24.505-64.648 24.303Q-64.944 24.101-64.944 23.780M-64.489 23.780Q-64.489 24.009-64.270 24.158Q-64.052 24.307-63.759 24.375Q-63.467 24.443-63.228 24.443Q-63.064 24.443-62.855 24.407Q-62.647 24.372-62.440 24.291Q-62.233 24.211-62.102 24.083Q-61.970 23.955-61.970 23.780Q-61.970 23.428-62.351 23.334Q-62.732 23.240-63.235 23.240L-63.853 23.240Q-64.093 23.240-64.291 23.391Q-64.489 23.541-64.489 23.780M-63.440 21.993Q-62.773 21.993-62.773 21.196Q-62.773 20.396-63.440 20.396Q-64.110 20.396-64.110 21.196Q-64.110 21.993-63.440 21.993M-60.962 21.712Q-60.962 21.391-60.837 21.102Q-60.712 20.813-60.487 20.590Q-60.261 20.366-59.966 20.246Q-59.670 20.126-59.352 20.126Q-59.024 20.126-58.762 20.226Q-58.501 20.325-58.325 20.507Q-58.149 20.690-58.055 20.948Q-57.961 21.206-57.961 21.538Q-57.961 21.630-58.043 21.651L-60.299 21.651L-60.299 21.712Q-60.299 22.300-60.015 22.683Q-59.731 23.066-59.164 23.066Q-58.843 23.066-58.574 22.873Q-58.306 22.680-58.217 22.365Q-58.210 22.324-58.135 22.310L-58.043 22.310Q-57.961 22.334-57.961 22.406Q-57.961 22.413-57.968 22.440Q-58.081 22.837-58.451 23.076Q-58.822 23.315-59.246 23.315Q-59.684 23.315-60.083 23.107Q-60.483 22.898-60.723 22.531Q-60.962 22.164-60.962 21.712M-60.292 21.442L-58.477 21.442Q-58.477 21.165-58.574 20.913Q-58.672 20.660-58.870 20.504Q-59.068 20.349-59.352 20.349Q-59.629 20.349-59.842 20.507Q-60.056 20.666-60.174 20.921Q-60.292 21.176-60.292 21.442M-56.847 22.406L-56.847 20.509L-57.486 20.509L-57.486 20.287Q-57.168 20.287-56.951 20.077Q-56.734 19.867-56.633 19.557Q-56.532 19.248-56.532 18.940L-56.266 18.940L-56.266 20.229L-55.189 20.229L-55.189 20.509L-56.266 20.509L-56.266 22.393Q-56.266 22.669-56.161 22.868Q-56.057 23.066-55.797 23.066Q-55.640 23.066-55.534 22.962Q-55.428 22.857-55.379 22.704Q-55.329 22.550-55.329 22.393L-55.329 21.979L-55.062 21.979L-55.062 22.406Q-55.062 22.632-55.162 22.842Q-55.261 23.052-55.445 23.184Q-55.630 23.315-55.859 23.315Q-56.296 23.315-56.571 23.078Q-56.847 22.840-56.847 22.406M-53.931 25.004L-53.999 25.004Q-54.034 25.004-54.056 24.978Q-54.078 24.953-54.078 24.918Q-54.078 24.874-54.047 24.857Q-53.692 24.553-53.442 24.163Q-53.193 23.773-53.041 23.341Q-52.889 22.909-52.819 22.440Q-52.749 21.972-52.749 21.497Q-52.749 21.018-52.819 20.552Q-52.889 20.085-53.042 19.650Q-53.196 19.214-53.447 18.826Q-53.699 18.438-54.047 18.144Q-54.078 18.127-54.078 18.082Q-54.078 18.048-54.056 18.023Q-54.034 17.997-53.999 17.997L-53.931 17.997Q-53.921 17.997-53.912 17.999Q-53.904 18-53.894 18.004Q-53.350 18.404-52.978 18.957Q-52.605 19.511-52.424 20.157Q-52.243 20.803-52.243 21.497Q-52.243 22.198-52.424 22.845Q-52.605 23.493-52.979 24.047Q-53.353 24.601-53.894 24.997Q-53.904 24.997-53.912 24.999Q-53.921 25-53.931 25.004\" fill=\"currentColor\" stroke=\"currentColor\" class=\"tikz-text\" style=\"stroke-width:0.210\"\u002F>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fg>\u003C\u002Fsvg>\u003Cfigcaption class=\"tikz-cap\">The encoder-decoder model. The encoder RNN compresses the whole source into a single context vector \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.4444em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathbf\">c\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>, which seeds the decoder RNN; the decoder generates the target autoregressively. Everything the decoder knows about the source must pass through \u003Cspan class=\"katex\">\u003Cspan class=\"katex-html\" aria-hidden=\"true\">\u003Cspan class=\"base\">\u003Cspan class=\"strut\" style=\"height:0.4444em;\">\u003C\u002Fspan>\u003Cspan class=\"mord mathbf\">c\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fspan>.\u003C\u002Ffigcaption>",1785117823393]