Regularized quasi-monotone method for stochastic optimization

DOI10.1007/S11590-022-01931-4MaRDI QIDQ6155646zbMATH OpenOpenAlexFDO

Authors

Publication date 5 June 2023

Published in Optimization Letters (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2107.03779

zbMATH Keywords

composite minimizationindividual convergencequasi-monotone methodregularizationstochastic optimization

Mathematics Subject Classification ID

Stochastic programming (90C15)

Abstract: We adapt the quasi-monotone method from [2] for composite convex minimization in the stochastic setting. For the proposed numerical scheme we derive the optimal convergence rate in terms of the last iterate, rather than on average as it is standard for subgradient methods. The theoretical guarantee for individual convergence of the regularized quasi-monotone method is confirmed by numerical experiments on l1-regularized robust linear regression.

Recommendations

• Stochastic subgradient method for quasi-convex optimization problems
• A Variable Sample-Size Stochastic Quasi-Newton Method for Smooth and Nonsmooth Stochastic Convex Optimization
• Quasi-monotone subgradient methods for nonsmooth convex minimization
• On convergence of quasi-gradient stochastic methods of optimization
• An optimal method for stochastic composite optimization

Cites work

• A parameterized proximal point algorithm for separable convex optimization
• Distributed optimization and statistical learning via the alternating direction method of multipliers
• Distributed quasi-monotone subgradient algorithm for nonsmooth convex optimization over directed graphs
• Dual averaging methods for regularized stochastic learning and online optimization
• Quasi-monotone subgradient methods for nonsmooth convex minimization
• The proximal alternating minimization algorithm for two-block separable convex optimization problems with linear constraints

Cited in

(3)

• scientific article; zbMATH DE number 3871036 (Why is no real title available?)
• The Stochastic Fejér-Monotone Hybrid Steepest Descent Method and the Hierarchical RLS
• The regularized stochastic Nesterov's accelerated quasi-Newton method with applications

This page was built for publication: Regularized quasi-monotone method for stochastic optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6155646)