Semi-flat minima and saddle points by embedding neural networks to overparameterization
Machine Learning
2019-06-17 v2 Machine Learning
Abstract
We theoretically study the landscape of the training error for neural networks in overparameterized cases. We consider three basic methods for embedding a network into a wider one with more hidden units, and discuss whether a minimum point of the narrower network gives a minimum or saddle point of the wider one. Our results show that the networks with smooth and ReLU activation have different partially flat landscapes around the embedded point. We also relate these results to a difference of their generalization abilities in overparameterized realization.
Cite
@article{arxiv.1906.04868,
title = {Semi-flat minima and saddle points by embedding neural networks to overparameterization},
author = {Kenji Fukumizu and Shoichiro Yamaguchi and Yoh-ichi Mototake and Mirai Tanaka},
journal= {arXiv preprint arXiv:1906.04868},
year = {2019}
}
Comments
38 pages, 4 figures