English

Stochastic Majorization-Minimization Algorithms for Large-Scale Optimization

Machine Learning 2013-09-11 v2 Machine Learning Optimization and Control

Abstract

Majorization-minimization algorithms consist of iteratively minimizing a majorizing surrogate of an objective function. Because of its simplicity and its wide applicability, this principle has been very popular in statistics and in signal processing. In this paper, we intend to make this principle scalable. We introduce a stochastic majorization-minimization scheme which is able to deal with large-scale or possibly infinite data sets. When applied to convex optimization problems under suitable assumptions, we show that it achieves an expected convergence rate of O(1/n)O(1/\sqrt{n}) after nn iterations, and of O(1/n)O(1/n) for strongly convex functions. Equally important, our scheme almost surely converges to stationary points for a large class of non-convex problems. We develop several efficient algorithms based on our framework. First, we propose a new stochastic proximal gradient method, which experimentally matches state-of-the-art solvers for large-scale 1\ell_1-logistic regression. Second, we develop an online DC programming algorithm for non-convex sparse estimation. Finally, we demonstrate the effectiveness of our approach for solving large-scale structured matrix factorization problems.

Keywords

Cite

@article{arxiv.1306.4650,
  title  = {Stochastic Majorization-Minimization Algorithms for Large-Scale Optimization},
  author = {Julien Mairal},
  journal= {arXiv preprint arXiv:1306.4650},
  year   = {2013}
}

Comments

accepted for publication for Neural Information Processing Systems (NIPS) 2013. This is the 9-pages version followed by 16 pages of appendices. The title has changed compared to the first technical report

R2 v1 2026-06-22T00:37:03.005Z