Toward Understanding the Importance of Noise in Training Neural Networks

[edit]

Mo Zhou, Tianyi Liu, Yan Li, Dachao Lin, Enlu Zhou, Tuo Zhao ;
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:7594-7602, 2019.

Abstract

Numerous empirical evidence has corroborated that the noise plays a crucial rule in effective and efficient training of deep neural networks. The theory behind, however, is still largely unknown. This paper studies this fundamental problem through training a simple two-layer convolutional neural network model. Although training such a network requires to solve a non-convex optimization problem with a spurious local optimum and a global optimum, we prove that a perturbed gradient descent algorithm in conjunction with noise annealing is guaranteed to converge to a global optimum in polynomial time with arbitrary initialization. This implies that the noise enables the algorithm to efficiently escape from the spurious local optimum. Numerical experiments are provided to support our theory.

Related Material