Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wang, Ruyu, Zhang, Chao
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:https://arxiv.org/abs/2308.01252
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912864401358848
author Wang, Ruyu
Zhang, Chao
author_facet Wang, Ruyu
Zhang, Chao
contents We propose a novel stochastic smoothing accelerated gradient (SSAG) method for general constrained nonsmooth convex composite optimization, and analyze the convergence rates. The SSAG method allows various smoothing techniques, and can deal with the nonsmooth term that is not easy to compute its proximal term, or that does not own the linear max structure. To the best of our knowledge, it is the first time to develop a stochastic approximation type method that treats the maximization of finite but numerous nonsmooth convex functions as a stochastic function, which significantly improves the computational efficiency. We prove that the SSAG method can simultaneously achieve the best-known order ${\cal{O}}(\frac{1}ε)$ of iteration complexity, and the optimal order ${\cal{O}}(\frac{1}{ε^2})$ of $\cal{SFO}$ complexity, using variable sample-size. Numerical results on the application arising from the distributionally robust optimization demonstrate the effectiveness and efficiency of the proposed SSAG method.
format Preprint
id arxiv_https___arxiv_org_abs_2308_01252
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Stochastic smoothing accelerated gradient method for general constrained nonsmooth convex composite optimization
Wang, Ruyu
Zhang, Chao
Optimization and Control
We propose a novel stochastic smoothing accelerated gradient (SSAG) method for general constrained nonsmooth convex composite optimization, and analyze the convergence rates. The SSAG method allows various smoothing techniques, and can deal with the nonsmooth term that is not easy to compute its proximal term, or that does not own the linear max structure. To the best of our knowledge, it is the first time to develop a stochastic approximation type method that treats the maximization of finite but numerous nonsmooth convex functions as a stochastic function, which significantly improves the computational efficiency. We prove that the SSAG method can simultaneously achieve the best-known order ${\cal{O}}(\frac{1}ε)$ of iteration complexity, and the optimal order ${\cal{O}}(\frac{1}{ε^2})$ of $\cal{SFO}$ complexity, using variable sample-size. Numerical results on the application arising from the distributionally robust optimization demonstrate the effectiveness and efficiency of the proposed SSAG method.
title Stochastic smoothing accelerated gradient method for general constrained nonsmooth convex composite optimization
topic Optimization and Control
url https://arxiv.org/abs/2308.01252