NearlyOptimalReward-FreeReinforcementLearningZihanZhang1SimonS.Du2XiangyangJi1AbstractRLisexplorationforwhichtheagentneedstostrategicallyvisitnewstatestolearntransitionandrewardinformationWestudyth...
AchievingNearInstance-OptimalityandMinimax-OptimalityinStochasticandAdversarialLiNearBanditsSimultaneouslyChung-WeiLee1HaipengLuo1Chen-YuWei1MengxiaoZhang1XiaojinZhang2Abstractadversarialenvironmen...
Sub-liNearMemorySketchesforNearNeighborSearchonStreamingDataBenjaminColeman1RichardGBaraniuk12AnshumaliShrivastava12Abstracthavehighsimilaritytoanydynamicallygeneratedqueryq.Wepresentthefirstsublin...
NearInputSparsityTimeKernelEmbeddingsviaAdaptiveSamplingDavidP.Woodruff1AmirZandieh2Abstractproducingcompressedandlow-rankapproximationstoker-nelmatrices(Rahimi&Recht,2008;Alaoui&Mahoney,Toaccelera...
LearningNearOptimalPolicieswithLowInherentBellmanErrorAndreaZanette1AlessandroLazaric2MykelKochenderfer1EmmaBrunskill1Abstract1.IntroductionWestudytheexplorationproblemwithapprox-Improvingthesample...
NearoptimalfinitetimeidentificationofarbitraryliNeardynamicalsystemsTuhinSarkar1AlexanderRakhlin2AbstractpopularliNearfeedbackcontrolsystemfoundinavarietyofdevices,fromplanetarysoftlandingsystemsfo...
NearOptimalFrequentDirectionsforSketchingDenseandSparseMatricesZengfengHuang1Abstractreceivedlotsofattentionrecently(Liberty,2013;Ghashami&Phillips,2014;Woodruff,2014;Ghashamietal.,2016;Givenalarge...