RegularizedSubmodularMaximizationatScaleEhsanKazemi1ShervinMinaee2MoranFeldman3AminKarbasi4Abstractpromotescoverageandrepresentativeness(Weietal.,2015)(wecallthisanoptimizationperspective)orsamplin...
LargeScalePrivateLearningviaLow-rankReparametrizationDaYu12HuishuaiZhang2WeiChen2JianYin1Tie-YanLiu2AbstractFigure1.Theproposedreparametrizationscheme.Theresidualweightmakesthereparametrizedoutputt...
NormalizedFlatMinima:ExploringScaleInvariantDefinitionofFlatMinimaforNeuralNetworksUsingPAC-BayesianAnalysisYusukeTsuzuku12IsseiSato12MasashiSugiyama21Abstractbleexplanationforthegeneralizationabil...
GreedyLayerwiseLearningCanScaletoImageNetEugeneBelilovsky1MichaelEickenberg2EdouardOyallon3Abstractasspeechrecognition(Chanetal.,2016),naturallanguageprocessing(Vaswanietal.,2017),andreinforcementl...
DataSummarizationatScale:ATwo-StageSubmodularApproachMarkoMitrovic1EhsanKazemi1MortezaZadimoghaddam2AminKarbasi1AbstractdriverssuggestedwaitinglocationsacrossNewYorkCitybasedonhistoricalriderpick-u...
LearnedOptimizersthatScaleandGeneralizeOlgaWichrowska1NiruMaheswaranathan23MatthewW.Hoffman4SergioGo´mezColmenarejo4MishaDenil4NandodeFreitas4JaschaSohl-Dickstein1Abstractlearningshowthat,givensuf...