English

On perturbed proximal gradient algorithms

Statistics Theory 2016-11-22 v4 Optimization and Control Statistics Theory

Abstract

We study a version of the proximal gradient algorithm for which the gradient is intractable and is approximated by Monte Carlo methods (and in particular Markov Chain Monte Carlo). We derive conditions on the step size and the Monte Carlo batch size under which convergence is guaranteed: both increasing batch size and constant batch size are considered. We also derive non-asymptotic bounds for an averaged version. Our results cover both the cases of biased and unbiased Monte Carlo approximation. To support our findings, we discuss the inference of a sparse generalized linear model with random effect and the problem of learning the edge structure and parameters of sparse undirected graphical models.

Keywords

Cite

@article{arxiv.1402.2365,
  title  = {On perturbed proximal gradient algorithms},
  author = {Yves F. Atchade and Gersende Fort and Eric Moulines},
  journal= {arXiv preprint arXiv:1402.2365},
  year   = {2016}
}

Comments

33 pages, 5 figures

R2 v1 2026-06-22T03:05:20.894Z