AlmostOptimalAnytimeAlgorithmforBatchedMulti-ArmedBanditsTianyuanJin1JingTang2PanXu3KekeHuang1XiaokuiXiao1QuanquanGu3Abstractittoguidethenextaction.However,thisisimpracticalformanyrealapplicationsw...
OrthogonalizedSGDandNestedArchitecturesforAnytimeNeuralNetworksChengchengWan1HenryHoffmann1ShanLu1MichaelMaire1Abstract...O1WeproposeanovelvariantofSGDcustomizedInput...O2fortrainingnetworkarchitec...
AnytimeOnline-to-Batch,OptimismandAccelerationAshokCutkosky1Abstractoptimalornear-optimalguarantees.ThishashelpedfuelthewidespreadadoptionofonlinelearningalgorithmsasAstandardwaytoobtainconvergence...