English

Variance-Reduced Proximal Stochastic Gradient Descent for Non-convex Composite optimization

Machine Learning 2016-09-13 v2 Machine Learning Numerical Analysis

Abstract

Here we study non-convex composite optimization: first, a finite-sum of smooth but non-convex functions, and second, a general function that admits a simple proximal mapping. Most research on stochastic methods for composite optimization assumes convexity or strong convexity of each function. In this paper, we extend this problem into the non-convex setting using variance reduction techniques, such as prox-SVRG and prox-SAGA. We prove that, with a constant step size, both prox-SVRG and prox-SAGA are suitable for non-convex composite optimization, and help the problem converge to a stationary point within O(1/ϵ)O(1/\epsilon) iterations. That is similar to the convergence rate seen with the state-of-the-art RSAG method and faster than stochastic gradient descent. Our analysis is also extended into the min-batch setting, which linearly accelerates the convergence. To the best of our knowledge, this is the first analysis of convergence rate of variance-reduced proximal stochastic gradient for non-convex composite optimization.

Keywords

Cite

@article{arxiv.1606.00602,
  title  = {Variance-Reduced Proximal Stochastic Gradient Descent for Non-convex Composite optimization},
  author = {Xiyu Yu and Dacheng Tao},
  journal= {arXiv preprint arXiv:1606.00602},
  year   = {2016}
}

Comments

This paper has been withdrawn by the author due to an error in the proof of the convergence rate. They will modify this proof as soon as possible

R2 v1 2026-06-22T14:15:43.570Z