Fine-GrainedAnalysisofStabilityandGeneralizationforStochasticGradientDescentYunwenLei12YimingYing3Abstractaniterativealgorithm,SGDupdatesthemodelsequentiallyuponreceivinganewdatumwithacheapper-iter...
FastLearningofGraphNeuralNetworkswithGuaranteedGeneralizability:One-hidden-layerCaseShuaiZhang1MengWang1SijiaLiu2Pin-YuChen3JinjunXiong3Abstract1.IntroductionAlthoughgraphneuralnetworks(GNNs)haveGr...
FastComputationofNashEquilibriainImperfectInformationGamesRemiMunos1JulienPerolat1Jean-BaptisteLespiau1MarkRowland1BartDeVylder1MarcLanctot1FinbarrTimbers1DanielHennes1ShayeganOmidshafiei1AudrunasG...
FastandConsistentLearningofHiddenMarkovModelsbyIncorporatingNon-ConsecutiveCorrelationsRobertMattila1CristianR.Rojas1EricMoulines23VikramKrishnamurthy4BoWahlberg1Abstractcomputationalbiology(Durbin...
Explore,DiscoverandLearn:UnsupervisedDiscoveryofState-CoveringSkillsV´ıctorCampos1AlexTrott2CaimingXiong2RichardSocher2XavierGiro-i-Nieto3JordiTorres1AbstractFigure1.Skillslearnedonamazewithbottl...
ExplainingGroupsofPointsinLow-DimensionalRepresentationsGregoryPlumb1JonathanTerhorst2SriramSankararaman3AmeetTalwalkar14AbstractFigure1:Arepresentationlearnedforasingle-cellRNAsequencedatasetusing...
EvaluatingthePerformanceofReinforcementLearningAlgorithmsScottM.Jordan1YashChandak1DanielCohen1MengxueZhang1PhilipS.Thomas1AbstractusabilityofRLalgorithms,wesuggestthatitshouldhavefourproperties.Fi...
EvaluatingLossyCompressionRatesofDeepGenerativeModelsSicongHuang123AlirezaMakhzani12YanshuaiCao3RogerGrosse12Abstracttance(FID)(Heuseletal.,2017),whichdonothavenearlythesamedegreeoftheoreticalunder...
EstimationofBoundsonPotentialOutcomesForDecisionMakingMaggieMakar1FredrikJohansson2JohnGuttag1DavidSontag1Abstractneedstoknowifthepotentialoutcomesundertreatmentandnon-treatmentfallwithin2–3.Learn...
EstimatingtheNumberandEffectSizesofNon-nullHypothesesJenniferBrennan1RamyaKorlakaiVinayak1KevinJamieson1AbstractForexample,supposeascientistwouldliketotest10,000Westudytheproblemofestimatingthedist...
EstimatingtheErrorofRandomizedNewtonMethods:ABootstrapApproachJessieX.T.Chen1MilesE.Lopes2AbstractresearchhasbeendrivenbythefactthatcomputinganinverseHessiantomachineprecisioncanbeverycostlyRandomi...
Error-BoundedCorrectionofNoisyLabelsSongzhuZheng1PengxiangWu2AmanGoswami3MayankGoswami4DimitrisMetaxas2ChaoChen5Abstract&GeoffreyE.,2012;Wuetal.,2018).Tocollectlargescaleannotateddata,itisin-Manyme...
EstimatingModelUncertaintyofNeuralNetworksinSparseInformationFormJongseokLee1MatthiasHumt1JianxiangFeng12RudolphTriebel12Abstractferencewhatlinearmapsareforalgebra.Forexample,oncetheposteriordistri...
EmpiricalStudyoftheBenefitsofOverparameterizationinLearningLatentVariableModelsRares-DariusBuhai1YoniHalpern2YoonKim3AndrejRisteski4DavidSontag1Abstractgoalisfrequentlytoconfirmorrejectwhetherapart...
EmergenceofSeparableManifoldsinDeepLanguageRepresentationsJonathanMamou1HangLe2MiguelADelRio2CoryStephenson1HanlinTang1YoonKim3SueYeonChung24Abstract1.IntroductionandRelatedWorkDeepneuralnetworks(D...
EliminatingtheInvarianceontheLossLandscapeofLinearAutoencodersRezaoftadeh1JiayiShen1ZhangyangWang1DylanShell1Abstractgenerallythecategorizationofextremepoints,hasflour-ished.Muchworkextendstheresul...
EinsumNetworks:FastandScalableLearningofTractableProbabilisticCircuitsRobertPeharz1StevenLang2AntonioVergari3KarlStelzner2AlejandroMolina2MartinTrapp4GuyVandenBroeck3KristianKersting2ZoubinGhahrama...
EfficientProximalMappingofthe1-path-normofShallowNetworksFabianLatore1PaulRolland1NadavHallak1VolkanCevher1AbstracttheLipschitzconstantofthenetworkfunctiontoquantizerobustness,andregularizationtoen...
DynamicsofDeepNeuralNetworksandNeuralTangentHierarchyJiaoyangHuang1Horng-TzerYau2Abstractproblems,thereisnoguaranteethatagradientbasedalgo-rithmwillbeabletofindtheoptimalparametersefficientlyTheevo...
DifferentiableLikelihoodsforFastInversionof‘Likelihood-Free’DynamicalSystemsHansKersting123NicholasKra¨mer1MartinSchiegg3ChristianDaniel3MichaelTiemann3PhilippHennig12AbstractFigure1.Inferenceon...