Improved Optimization of Finite Sums with Minibatch Stochastic Variance Reduced Proximal Iterations
Optimization and Control
2017-10-12 v2 Machine Learning
Machine Learning
Abstract
We present novel minibatch stochastic optimization methods for empirical risk minimization problems, the methods efficiently leverage variance reduced first-order and sub-sampled higher-order information to accelerate the convergence speed. For quadratic objectives, we prove improved iteration complexity over state-of-the-art under reasonable assumptions. We also provide empirical evidence of the advantages of our method compared to existing approaches in the literature.
Keywords
Cite
@article{arxiv.1706.07001,
title = {Improved Optimization of Finite Sums with Minibatch Stochastic Variance Reduced Proximal Iterations},
author = {Jialei Wang and Tong Zhang},
journal= {arXiv preprint arXiv:1706.07001},
year = {2017}
}