Adaptive Learning Rate via Covariance Matrix Based Preconditioning for Deep Neural Networks
Machine Learning
2017-09-29 v2 Artificial Intelligence
Machine Learning
Abstract
Adaptive learning rate algorithms such as RMSProp are widely used for training deep neural networks. RMSProp offers efficient training since it uses first order gradients to approximate Hessian-based preconditioning. However, since the first order gradients include noise caused by stochastic optimization, the approximation may be inaccurate. In this paper, we propose a novel adaptive learning rate algorithm called SDProp. Its key idea is effective handling of the noise by preconditioning based on covariance matrix. For various neural networks, our approach is more efficient and effective than RMSProp and its variant.
Cite
@article{arxiv.1605.09593,
title = {Adaptive Learning Rate via Covariance Matrix Based Preconditioning for Deep Neural Networks},
author = {Yasutoshi Ida and Yasuhiro Fujiwara and Sotetsu Iwamura},
journal= {arXiv preprint arXiv:1605.09593},
year = {2017}
}
Comments
Accepted at IJCAI 2017