TheHintonsinYourNeuralNetwork:aQuantumFieldTheoryViewofDeepLearningRobertoBondesan1MaxWelling1AbstractInthispaperweproposeadirectmappingofadeepneu-ralnetworkontoaopticalquantumcomputerthroughtheInt...
HowtoLearnwhenDataReactstoYourModel:PerformativeGradientDescentZacharyIzzo1LexingYing12JamesZou3Abstractimagingdatafromonehospitalisdeployedatadifferenthospitalduetothedifferenceinimagingdevices.Ti...
HowtoTrainYourNeuralODE:theWorldofJacobianandKineticRegularizationChrisFinlay1Jo¨rn-HenrikJacobsen2LevonNurbekyan3AdamMOberman1Abstract(a)Optimaltransportmap(b)genericflowTrainingneuralODEsonlarge...
Don’tWasteYourBits!SqueezeActivationsandGradientsforDeepNeuralNetworksviaTINYSCRIPTFangchengFu12YuzhengHu1YihanHe1JiaweiJiang3YingxiaShao4CeZhang3BinCui15Abstractdrawback1:Recentyearshavewitnessed...
LearnfromYourNeighbor:LearningMulti-modalMappingsfromSparseAnnotationsAshwinKalyan1StefanLee1AnithaKannan2DhruvBatra13AbstractAhomekitchenwithwoodencabinetsandadinningtable.Manystructuredprediction...