Stochastic Conjugate Gradient Algorithm With Variance Reduction

Xiaobo Jin

doi:10.1109/TNNLS.2018.2868835

Stochastic Conjugate Gradient Algorithm With Variance Reduction

2019-05-01

39

OA

AI

DOI：10.1109/TNNLS.2018.2868835

Original

Save

View PDF

Abstract

En 中文

Conjugate gradient (CG) methods are a class of important methods for solving linear equations and nonlinear optimization problems. In this paper, we propose a new stochastic CG algorithm with variance reduction(1) and we prove its linear convergence with the Fletcher and Reeves method for strongly convex and smooth functions. We experimentally demonstrate that the CG with variance reduction algorithm converges faster than its counterparts for four learning models, which may be convex, nonconvex or nonsmooth. In addition, its area under the curve performance on six large-scale data sets is comparable to that of the LIBLINEAR solver for the L2-regularized L2-loss but with a significant improvement in computational efficiency.

Keywords:

Computational efficiency

covariance reduction

empirical risk minimization (ERM)

linear convergence

stochastic conjugate gradient (CG)

Stochastic Conjugate Gradient Algorithm With Variance Reduction

Abstract

AI Summary

Journal

Organization