Login / Signup
Why Does Large Batch Training Result in Poor Generalization? A Comprehensive Explanation and a Better Strategy from the Viewpoint of Stochastic Optimization.
Tomoumi Takase
Satoshi Oyama
Masahito Kurihara
Published in:
Neural Comput. (2018)
Keyphrases
</>
stochastic optimization
viewpoint
multistage
batch mode
decision making
cooperative
training set
special case
training examples
training process
np hard
probability distribution