Randomized Smoothing SVRG for Large-scale Nonsmooth Convex Optimization
Abstract
In this paper, we consider the problem of minimizing the average of a large number of nonsmooth and convex functions. Such problems often arise in typical machine learning problems as empirical risk minimization, but are computationally very challenging. We develop and analyze a new algorithm that achieves robust linear convergence rate, and both its time complexity and gradient complexity are superior than state-of-art nonsmooth algorithms and subgradient-based schemes. Besides, our algorithm works without any extra error bound conditions on the objective function as well as the common strongly-convex condition. We show that our algorithm has wide applications in optimization and machine learning problems, and demonstrate experimentally that it performs well on a large-scale ranking problem.
Cite
@article{arxiv.1805.05189,
title = {Randomized Smoothing SVRG for Large-scale Nonsmooth Convex Optimization},
author = {Wenjie Huang},
journal= {arXiv preprint arXiv:1805.05189},
year = {2018}
}
Comments
10 pages, 12 figures. arXiv admin note: text overlap with arXiv:1103.4296, arXiv:1403.4699 by other authors