ProvableMeta-LearningofLinearRepresentationsNileshTripuraneni1ChiJin2MichaelI.Jordan1Abstractrangingfromdeepreinforcementlearning(Baevskietal.,2019)tonaturallanguageprocessing(Ando&Zhang,2005;Meta-...
PACOH:Bayes-OptimalMeta-LearningwithPAC-GuaranteesJonasRothfuss1VincentFortuin1MartinJosifoski2AndreasKrause1Abstractmeta-trainingtargettrainingtargettestingMeta-Learningcansuccessfullyacquireusefu...
MURAL:Meta-LearningUncertainty-AwareRewardsforOutcome-DrivenReinforcementLearningKevinLi1AbhishekGupta1VitchyrPong1AshwinReddy1AurickZhou1JustinYu1SergeyLevine1AbstractFigure1.MURAL:Ourmethodtrains...
Meta-LearningBidirectionalUpdateRulesMarkSandler1MaxVladymyrov1AndreyZhmoginov1NolanMiller1AndrewJackson1TomMadams1BlaiseAgu¨erayArcas1AbstracttracedbacktoRobbins&Monro(1951).Sinceitsintro-duction...
Meta-LearningHyperparameterPerformancePredictionwithNeuralProcessesYingWei1PeilinZhao2JunzhouHuang2AbstractFreefromthelimitationsofexhaustivesearch(Bergstra&Bengio,2012)beingcomputationallyexpensiv...
ImprovingGeneralizationinMeta-LearningviaTaskAugmentationHuaxiuYao†1Long-KaiHuang2LinjunZhang3YingWei4LiTian2JamesZou1JunzhouHuang2ZhenhuiLi5Abstractquicklylearninganewtaskwithaslittleasafewexampl...
Large-ScaleMeta-LearningwithContinualTrajectoryShiftingJaeWoongShin1HaeBeomLee1BoqingGong21SungJuHwang13Abstractexamples(Lakeetal.,2015;Vinyalsetal.,2016;Santoroetal.,2016;Snelletal.,2017;Finnetal....
DataAugmentationforMeta-LearningRenkunNi1MicahGoldblum1AmrSharaf2KezhiKong1TomGoldstein1AbstractMeta-Learningframeworksusedataformultiplepurposesduringeachgradientupdate,whichcreatesthepossibilityf...
BridgingMulti-TaskLearningandMeta-Learning:TowardsEfficientTrainingandEffectiveAdaptationHaoxiangWang1HanZhao1BoLi1Abstract1.IntroductionMulti-tasklearning(MTL)aimstoimprovetheMulti-tasklearninghas...
UnravelingMeta-Learning:UnderstandingFeatureRepresentationsforFew-ShotTasksMicahGoldblum1StevenReich1LiamFowl1RenkunNi1ValeriiaCherepanova1TomGoldstein1Abstractdownentirelywhenonlyafewdatasamplesar...
TASKNORM:RethinkingBatchNormalizationforMeta-LearningJohnBronskill1JonathanGordon1JamesRequeima12SebastianNowozin3RichardE.Turner13Abstractthe-artperformanceinarangeofbenchmarktasks(Finnetal.,2017;...
OntheGlobalOptimalityofModel-AgnosticMeta-Learning:ReinforcementLearningandSupervisedLearningLingxiaoWang1QiCai1ZhuoyanYang2ZhaoranWang1Abstracttimization,wheretheinnerlevelsolveseachsubtaskbasedon...
Meta-LearningwithSharedAmortizedVariationalInferenceEkaterinaIakovleva1JakobVerbeek2KarteekAlahari1Abstract2016),nearestclass-meanapproaches(Dvorniketal.,2019;Mensinketal.,2012;Renetal.,2018;Snelle...
Meta-LearningwithStochasticLinearBanditsLeonardoCella12AlessandroLazaric3MassimilianoPontil2AbstractsolidatedMABsettinginwhicheacharmisassociatedwithavectoroffeaturesandthearmpayofffunctionismod-We...
Meta-LearningforMixedLinearRegressionWeihaoKong1RaghavSomani1ZhaoSong2ShamKakade1SewoongOh1Abstractexamples(Wangetal.,2017).Inmodernsupervisedlearning,therearealargeInspiredbyhumaningenuityinquickl...
MetaFun:Meta-LearningwithIterativeFunctionalUpdatesJinXu1Jean-FrancoisTon1HyunjikKim2AdamR.Kosiorek13YeeWhyeTeh1AbstractTheobjectiveofMeta-Learningisthentolearntheencoderandthedecodersuchthatthepro...
Few-shotRelationExtractionviaBayesianMeta-LearningonRelationGraphsMengQu12TianyuGao3Louis-PascalA.C.Xhonneux12JianTang145Abstracttomaticallygeneratetrainingdata.Forexample,givenatriplet(Washington,...
ProvableGuaranteesforGradient-BasedMeta-LearningMikhailKhodak1Maria-FlorinaBalcan1AmeetTalwalkar12Abstractwithexamples(xi,yi)∈X×Yforanunseentask,thepopularMAMLalgorithm(Finnetal.,2017)outputsWest...
OnlineMeta-LearningChelseaFinn1AravindRajeswaran2ShamKakade2SergeyLevine1Abstractsettingwheretasksarerevealedoneafteranother,butaimstoattainzero-shotgeneralizationwithoutanytask-specificAcentralcap...
Meta-LearningNeuralBloomFiltersJackWRae12SergeyBartunov1TimothyPLillicrap12Abstracttives.OnecanfindBloomFiltersembeddedwithinawiderangeofproductionsystems;fromnetworksecurity(Ger-Therehasbeenarecen...