pith. sign in

arxiv: 1802.06502 · v3 · pith:KJLMSZFBnew · submitted 2018-02-19 · 💻 cs.LG

EA-CG: An Approximate Second-Order Method for Training Fully-Connected Neural Networks

classification 💻 cs.LG
keywords methodapproximatefcnnsmatrixtrainingapproximationcg-basedfully-connected
0
0 comments X
read the original abstract

For training fully-connected neural networks (FCNNs), we propose a practical approximate second-order method including: 1) an approximation of the Hessian matrix and 2) a conjugate gradient (CG) based method. Our proposed approximate Hessian matrix is memory-efficient and can be applied to any FCNNs where the activation and criterion functions are twice differentiable. We devise a CG-based method incorporating one-rank approximation to derive Newton directions for training FCNNs, which significantly reduces both space and time complexity. This CG-based method can be employed to solve any linear equation where the coefficient matrix is Kronecker-factored, symmetric and positive definite. Empirical studies show the efficacy and efficiency of our proposed method.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.