ImprovingGradientRegularizationusingComplex-ValuedNeuralNetworksEricYeats1YiranChen1HaiLi1Abstractandthattheseinputsaredifficulttodistinguishfromnor-malinputs.SubsequentinvestigationbyGoodfelloweta...
GradientDisaggregation:BreakingPrivacyinFederatedLearningbyReconstructingtheUserParticipantMatrixMaximilianLam1Gu-YeonWei1DavidBrooks1VijayJanapaReddi1MichaelMitzenmacher1Abstractorwearables,andmai...
GRAD-MATCH:GradientMatchingbasedDataSubsetSelectionforEfficientDeepModelTrainingKrishnatejaKillamsetty1DurgaSivasubramanian2GaneshRamakrishnan2AbirDe2RishabhIyer1Abstractlossintestaccuracy?Thegreat...
GlobalConvergenceofPolicyGradientforLinear-QuadraticMean-FieldControl/GameinContinuousTimeWeichenWang1JiequnHan2ZhuoranYang3ZhaoranWang4Abstractmorerealisticreal-worldproblems,suchasroboticcontrol(...
GeneralizedDoubly-ReparameterizedGradientEstimatorsMatthiasBauer1AndriyMnih1Abstractusuallypreferunbiasedestimatorsastheytendtobebetter-behavedandarebetterunderstood.LowervarianceisalsoEfficientlow...
GBHT:GradientBoostingHistogramTransformforDensityEstimationJingyiCui1HanyuanHang2YisenWang1ZhouchenLin13Abstractovertwodecades(Bühlmann&Yu,2003),boostingattractsmoreandmoreattentioninresearchesonm...
FastStochasticBregmanGradientMethods:SharpAnalysisandVarianceReductionRadu-AlexandruDragomir12MathieuEven23HadrienHendrikx23Abstractgoalistorecoveranunknownsignalfromalargenumberofnoisyobservations...
DecentralizedRiemannianGradientDescentontheStiefelManifoldShixiangChen1AlfredoGarcia1MingyiHong2ShahinShahrampour1AbstractagentoptimizationproblemWeconsiderdistributednon-convexoptimization1nwherea...
DatasetDynamicsviaGradientFlowsinProbabilitySpaceDavidAlvarez-Melis1Nicolo`Fusi1Abstract2019),datasetpooling,orinstancemixing(Zhangetal.,2018).Ontheotherhand—andourprincipalmotivationinVariousmach...
AgnosticLearningofHalfspaceswithGradientDescentviaSoftMarginsSpencerFrei1YuanCao2QuanquanGu2AbstractminimizethesurrogateriskWeanalyzethepropertiesofGradientdescentonF(w):=E(x,y)∼D(ywx).(1)convexsu...
AccumulatedDecoupledLearningwithGradientStalenessMitigationforConvolutionalNeuralNetworksHuipingZhuang1ZhenyuWeng1FulinLuo1Kar-AnnToh2HaizhouLi3ZhipingLin1Abstractefficiency,thedecoupledlearning(Ja...
LearningUnderDelayedFeedback:ImplicitlyAdaptingtoGradientDelaysRotemZamirAviv1IdoHakimi2AssafSchuster2KfirY.Levy13Abstractceed.Suchmethodsareeasiertoanalyze,yettheirper-formancedependsontheslowestm...
ARMS:Antithetic-REINFORCE-Multi-SampleGradientforBinaryVariablesAlekDimitriev1MingyuanZhou1AbstractIfbisdiscrete,however,Tφisnotdifferentiable,andfisnotalwaysdifferentiable,e.g.,ifitcorrespondstot...
AGradientBasedStrategyforHamiltonianMonteCarloHyperparameterOptimizationAndrewCampbell1WenlongChen2VincentStimper34Jose´MiguelHerna´ndez-Lobato3YichuanZhang5Abstractbehaviourandhighlycorrelatedsa...
APolicyGradientAlgorithmforLearningtoLearninMultiagentReinforcementLearningDong-KiKim12MiaoLiu23MatthewRiemer23ChuangchuangSun12MarwaAbdulhai12GolnazHabibi12SebastianLopez-Cot12GeraldTesauro23Jonat...
ANovelSequentialCoresetMethodforGradientDescentAlgorithmsJiaweiHuang1RuominHuang2WenjieLiu1NikolaosM.Freris1HuDing1AbstractoriginalsetP;thatis,wecanreplacePbyP˜whenrunningAwiderangeofoptimizationp...
TheImplicitRegularizationofStochasticGradientFlowforLeastSquaresAlnurAli1EdgarDobriban2RyanJ.Tibshirani3Abstractgeneratedbysequentialoptimizationalgorithmsmayserveascheapapproximationstothemoreexpe...
StochasticGradientandLangevinProcessesXiangCheng1DongYin1PeterBartlett1MichaelJordan1AbstractrestrictedtoisotropicGaussiannoise,whereasthenoiseinSGDcanbefarfromGaussian.Second,thenoisedependsWeprov...
StochasticHamiltonianGradientMethodsforSmoothGamesNicolasLoizou1HugoBerard12AlexiaJolicoeur-Martineau1PascalVincent†12SimonLacoste-Julien†1IoannisMitliagkas†1Abstractforeveryx1∈Rd1andx2∈Rd2.We...
StatisticallyPreconditionedAcceleratedGradientMethodforDistributedOptimizationHadrienHendrikx1LinXiao2Se´bastienBubeck2FrancisBach1LaurentMassoulie´1Abstractleaveψmainlyfornon-smoothregularizati...