BeyondUCB:OptimalandEfficientContextualBanditswithRegressionOraclesDylanJ.Foster1AlexanderRakhlin1Abstractible,generalpurposealgorithmsthatworkforarbitrary,user-specifiedclassesofpoliciesandcomewit...
TamingMAML:EfficientUnbiasedMeta-ReinforcementLearningHaoLiu1RichardSocher1CaimingXiong1Abstractreinforcementlearning(Meta-RL)(Wangetal.,2016;Duanetal.,2016;Mishraetal.,2018;Finnetal.,2017;NicholWh...
ProvablyefficientRLwithRichObservationsviaLatentStateDecodingSimonS.Du1AkshayKrishnamurthy2NanJiang3AlekhAgarwal4MiroslavDud´ık2JohnLangford2Abstract2010;Lattimore&Hutter,2012).Consequently,treat...
ProvablyEfficientImitationLearningfromObservationAloneWenSun1AnirudhVemula1ByronBoots2J.AndrewBagnell3Abstractaction,viasupervisedlearningapproaches(e.g.,DAgger(Rossetal.,2011),AggreVaTe(Ross&Bagne...
ProvablyEfficientMaximumEntropyExplorationEladHazan12ShamM.Kakade342KaranSingh12AbbyVanSoest12Abstractsuchaslearningwithintrinsicrewardandcuriositydrivenmethods,surveyedbelow.Ourworkstudiesaclassof...
PopulationBasedAugmentation:EfficientLearningofAugmentationPolicySchedulesDanielHo12EricLiang1IonStoica1PieterAbbeel13XiChen13AbstractBaselineAutoAugmentPopulationBasedAugmentation4Akeychallengeinl...
ParsimoniousBlack-BoxAdversarialAttacksviaEfficientCombinatorialOptimizationSeungyongMoon12GaonAn12HyunOhSong12AbstractHowever,inmorerealisticsettings(Watsonvisualrecogni-tion,2019;GooglevisionAPI,...
ParameterEfficientTrainingofDeepConvolutionalNeuralNetworksbyDynamicSparseReparameterizationHeshamMostafa1XinWang12Abstract1.IntroductionModerndeepneuralnetworksaretypicallyhighlyDeepneuralnetworks...
OnScalableandEfficientComputationofLargeScaleOptimalTransportYujiaXie1MinshuoChen1HaomingJiang1TuoZhao1HongyuanZha1AbstractearliestliteraturedatesbacktoMonge(1781).Sincethen,ithasattractedincreasin...
OnEfficientOptimalTransport:AnAnalysisofGreedyandAcceleratedMirrorDescentAlgorithmsTianyiLin1NhatHo2MichaelI.Jordan3Abstractscalableposteriorsamplingforlargedatasets(Srivastavaetal.,2015;2018),opti...
MoreEfficientOff-PolicyEvaluationthroughRegularizedTargetedLearningAure´lienF.Bibaut1IvanaMalenica1NikosVlassis2MarkJ.vanderLaan1Abstractinference,andhasledtomanymethodologicaldevelop-ments.Oneoft...
LegoNet:EfficientConvolutionalNeuralNetworkswithLegoFiltersZhaohuiYang12YunheWang2HantingChen12ChuanjianLiu2BoxinShi34ChaoXu1ChunjingXu2ChangXu5AbstractHuangetal.,2017),detection(Renetal.,2015;Liue...
LatentGNN:LearningEfficientNon-localRelationsforVisualRecognitionSongyangZhang1ShipengYan1XumingHe1Abstractvisioncommunitybylearninghierarchicalfeaturerepresen-tationsofimageswithdeepconvolutionaln...
ExploitingStructureofUncertaintyforEfficientMatroidSemi-BanditsPierrePerrault12VianneyPerchet23MichalValko14Abstracttimeandspacecomplexity)oftheircombinatorialgeneral-ization,inwhichtheagentselects...
EfficientOptimizationofLoopsandLimitswithRandomizedTelescopingSumsAlexBeatson1RyanP.Adams1Abstractovertime,eachlearningsteprequiresloopingovertimesteps.Morebroadly,inmanyscientificandengineeringWec...
EfficientOn-DeviceModelsusingNeuralProjectionsSujithRavi1Abstractofdatatolearnusefulrepresentationsthatcanbeusedtopredictoutputsatinferencetime.Forefficiencyrea-Manyapplicationsinvolvingvisualandla...
EfficientTrainingofBERTbyProgressivelyStackingLinyuanGong1DiHe1ZhuohanLi1TaoQin2LiweiWang13Tie-YanLiu2Abstractespeciallyindomainsthatrequireparticularexpertise.Unsupervisedpre-trainingiscommonlyuse...
EfficientOff-PolicyMeta-ReinforcementLearningviaProbabilisticContextVariablesKateRakelly1AurickZhou1DeirdreQuillen1ChelseaFinn1SergeyLevine1AbstractFortunately,manyoftheproblemswewouldlikeourau-ton...
EfficientNonconvexRegularizedTensorCompletionwithStructure-awareProximalIterationsQuanmingYao12JamesT.Kwok2BoHan3AbstractInmanyapplications,onlyafewtensorentriesareob-served.Forexample,ofteneachYou...
EfficientlearningofsmoothprobabilityfunctionsfromBernoullitestswithguaranteesPaulRolland1AliKavis1AlexImmer1AdishSingla2VolkanCevher1Abstractgrowscubiclywiththenumberoftests,andcanthusbeinapplicabl...