AdaScale SGD A User-Friendly Algorithm for Distributed Training

AdaScaleSGD:AUser-FriendlyAlgorithmforDistributedTrainingTylerB.Johnson†1PulkitAgrawal†1HaijieGu1CarlosGuestrin1Abstracttrainingalgorithms.Duringeachiteration,SGDappliesasmallandnoisyupdatetothem...