DebiasingModelUpdatesforImprovingPersonalizedFederatedTrainingDurmusAlpEmreAcar1YueZhao2RuizhaoZhu1RamonMatasNavarro2MatthewMattina2PaulN.Whatmough2VenkateshSaligrama1Abstractofdevicesparticipating...
MetaFun:Meta-LearningwithIterativeFunctionalUpdatesJinXu1Jean-FrancoisTon1HyunjikKim2AdamR.Kosiorek13YeeWhyeTeh1AbstractTheobjectiveofmeta-learningisthentolearntheencoderandthedecodersuchthatthepro...
MultiplicativeWeightsUpdateasaDistributedConstrainedOptimizationAlgorithm:ConvergencetoSecond-orderStationaryPointsAlmostAlwaysIoannisPanageas1GeorgiosPiliouras1XiaoWang1Abstracttion).Moreoverbyadd...
StingyCD:SafelyAvoidingWastefulUpdatesinCoordinateDescentTylerB.Johnson1CarlosGuestrin1Abstractrecentyears.Thisincludestheoretical(Nesterov,2012;Shalev-Shwartz&Tewari,2011)andmoreapplied(FanCoordin...
CombiningModel-BasedandModel-FreeUpdatesforTrajectory-CentricReinforcementLearningYevgenChebotar12KarolHausman1MarvinZhang3GauravSukhatme1StefanSchaal12SergeyLevine3AbstractFigure1.Realrobottasksus...