Stochastic Nonconvex Optimization with Large Minibatches
[edit]
Proceedings of the 30th International Conference on Algorithmic Learning Theory, PMLR 98:857882, 2019.
Abstract
We study stochastic optimization of nonconvex loss functions, which are typical objectives for training neural networks. We propose stochastic approximation algorithms which optimize a series of regularized, nonlinearized losses on large minibatches of samples, using only firstorder gradient information. Our algorithms provably converge to an approximate critical point of the expected objective with faster rates than minibatch stochastic gradient descent, and facilitate better parallelization by allowing larger minibatches.
Related Material


