OntheDifficultyofUnbiasedAlphaDivergenceMinimizationTomasGeffner1JustinDomke1AbstractExistingalpha-Divergenceminimizationalgorithmscanbeclassifiedintotwobroadgroups:biasedmethods(Li&Severalapproxim...
OfflineReinforcementLearningwithFisherDivergenceCriticRegularizationIlyaKostrikov12JonathanTompson2RobFergus13OfirNachum2Abstractwheredeployinganewpolicytointeractwiththeliveen-vironmentisexpensive...
Non-NegativeBregmanDivergenceMinimizationforDeepDirectDensityRatioEstimationMasahiroKato1TakeshiTeshima2Abstractthedensityratiowhenanomaly-freeunlabeledtestdataareavailable(Hidoetal.,2008).Densityr...
ImprovedContrastiveDivergenceTrainingofEnergy-BasedModelsYilunDu1ShuangLi1JoshuaTenenbaum1IgorMordatch2AbstractFigure1:(Left)128x128samplesonunconditionalCelebA-HQ.(Right)128x128samplesonunconditio...
SpreadDivergenceMingtianZhang1PeterHayes1TomBird1RazaHabib1DavidBarber1Abstractsee(Csiszár,1967;Ali&Silvey,1966).Awell-knownf-DivergenceistheKullback-Leibler(KL)DivergenceFordistributionsPandQwith...
ModelFusionwithKullback–LeiblerDivergenceSebastianClaici12MikhailYurochkin23SoumyaGhosh23JustinSolomon12Abstract(a)BayesianneuralnetworktrainedonbiaseddataWeproposeamethodtofuseposteriordistribu-t...
DeepDivergenceLearningKubraCilingir1RachelManzelli1BrianKulis1Abstractclassification(Davisetal.,2007;Weinberger&Saul,2009;Goldbergeretal.,2004).Thesemethods,knownasMa-Classicallinearmetriclearningm...
ComposingEntropicPoliciesusingDivergenceCorrectionJonathanJHunt1AndreBarreto1TimothyPLillicrap1NicolasHeess1Abstractetal.,2012;Haith&Krakauer,2013)However,oncesuchskillshavebeenacquiredhumansrapidl...
AContrastiveDivergenceforCombiningVariationalInferenceandMCMCFranciscoJ.R.Ruiz12MichalisK.Titsias3AbstractWedevelopamethodforcombiningVIandMCMCthatimprovesanexplicitvariationaldistribution(i.e.,wit...
ADivergenceBoundforHybridsofMCMCandVariationalInferenceandanApplicationtoLangevinDynamicsandSGVIJustinDomke1Abstractthem.Computingp(z)thusrequiresafullpassoverthedataset.TheideaofStochasticGradient...