CombiningPessimismwithOptimismforRobustandEfficientModel-BasedDeepReinforcementLearningSebastianCuri1IlijaBogunovic1AndreasKrause1Abstractunpredictableways.Themaingoalisthentolearnapolicythatprovab...
Communication-EfficientDistributedOptimizationwithQuantizedPreconditionersFoivosAlimisis1PeterDavies2DanAlistarh23AbstractInthispaper,wefocusonthecommunication(bit)com-plexityoftheclassicempiricalr...
CombinatorialBlockingBanditswithStochasticDelaysAlexiaAtsidakou1OrestisPapadigenopoulos2SoumyaBasu3ConstantineCaramanis1SanjayShakkottai1AbstractCella&Cesa-Bianchi,2019).Thesevariantscaptureappli-c...
CollaborativeBayesianOptimizationwithFairRegretRachaelHweeLingSim1YehongZhang2BryanKianHsiangLow1PatrickJaillet3Abstractperformancebysequentiallyselectinginputqueriesforeval-uatingtheobjectivefunct...
ClassificationwithRejectionBasedonCost-sensitiveClassificationNontawatCharoenphakdee12ZhenghangCui12YivanZhang12MasashiSugiyama21Abstractsificationincriticalapplications.Thegoalofclassificationwith...
Class2Simi:ANoiseReductionPerspectiveonLearningwithNoisyLabelsSonghuaWu1XiaoboXia1TongliangLiu1BoHan2MingmingGong3NannanWang4HaifengLiu5GangNiu6Abstractdealwiththelabelnoiseprobleminpointwisemanner...
CATE:Computation-awareNeuralArchitectureEncodingwithTransformersShenYan1KaiqiangSong23FeiLiu2MiZhang1Abstract2020)ordesigningefficientarchitecturesearchandevalu-ationmethods(Luoetal.,2018b;Shietal....
Byzantine-ResilientHigh-DimensionalSGDwithLocalIterationsonHeterogeneousDataDeepeshData1SuhasDiggavi1Abstract(Deanetal.,2012)(e.g.,trainingamachinelearningmodelwithoutcollectingtheclients’data,whi...
ConViT:ImprovingVisionTransformerswithSoftConvolutionalInductiveBiasesSte´phaned’Ascoli12HugoTouvron2MatthewL.Leavitt2AriS.Morcos2GiulioBiroli12LeventSagun2Abstract1.IntroductionConvolutionalarch...
ControllingGraphDynamicswithReinforcementLearningandGraphNeuralNetworksEliA.Meirom1HaggaiMaron1ShieMannor1GalChechik1AbstractFigure1.Aviralinfectionprocessonagraphandaninterventionaimedtostopitsspr...
BayesianOptimisticOptimisationwithExponentiallyDecayingRegretHungTran-The1SunilGupta1SantuRana1SvethaVenkatesh1Abstracttransformaglobaloptimisationproblemintoasequenceofauxiliaryoptimisationproblem...
BatchValue-functionApproximationwithOnlyRealizabilityTengyangXie1NanJiang1Abstractthissubproblem,wecreateapiecewiseconstantfunctionclassofstatisticalcomplexityO(1/2)thatcanexpressbothWemakeprogress...
BANG:BridgingAutoregressiveandNon-autoregressiveGenerationwithLargeScalePretrainingWeizhenQi12YeyunGong3JianJiao4YuYan4WeizhuChen4DayihengLiu25KewenTang4HouqiangLi1JiushengChen4RuofeiZhang4MingZhou...
Average-RewardOff-PolicyPolicyEvaluationwithFunctionApproximationShangtongZhang1YiWan2RichardS.Sutton2ShimonWhiteson1Abstractwhichaimtogenerateapolicythatmaximizestherewardratebyiterativelyimprovin...
AutomaticVariationalInferencewithCascadingFlowsLucaAmbrogioni1GianluigiSilvestri12MarcelvanGerven1AbstractKerstingandDeRaedt,2007;Pfeffer,2001;Parketal.,2005;Goodmanetal.,2012;Wingateetal.,2011;Pat...
BreakingtheDeadlyTriadwithaTargetNetworkShangtongZhang1HengshuaiYao23ShimonWhiteson1Abstractpingmethodsconstructupdatetargetsforanestimatebyusingtheestimateitselfrecursively,whichusuallyhaslowerThe...
AlphaNet:ImprovedTrainingofSupernetswithAlpha-DivergenceDilinWang1ChengyueGong2MengLi1QiangLiu2VikasChandra1AbstractNeuralarchitecturesearch(NAS)automatestheneuralnet-workdesignbyexploringanenormou...
AgnosticLearningofHalfspaceswithGradientDescentviaSoftMarginsSpencerFrei1YuanCao2QuanquanGu2AbstractminimizethesurrogateriskWeanalyzethepropertiesofgradientdescentonF(w):=E(x,y)∼D(ywx).(1)convexsu...
AdversarialPurificationwithScore-basedGenerativeModelsJongminYoon1SungJuHwang12JuhoLee12Abstract2019),inwhichaclassifieristrainedwithadversarialexam-ples,isconsideredasastandarddefensemethodduetoit...
AdversarialCombinatorialBanditswithGeneralNon-linearRewardFunctionsXiChen1YanjunHan2YiningWang3Abstractchoosesarewardvectorvt=(vt1,···,vtN)∈[0,1]Nnotrevealedtothealgorithm.Thealgorithmchoosesas...