Robins-Monro Augmented Lagrangian Method for Stochastic Convex Optimization

arXiv2208.14019MaRDI QIDQ6409211FDOQ6409211

Authors: Rui Wang, Chao Ding 

Publication date: 30 August 2022

Abstract: In this paper, we propose a Robbins-Monro augmented Lagrangian method (RMALM) to solve a class of constrained stochastic convex optimization, which can be regarded as a hybrid of the Robbins-Monro type stochastic approximation method and the augmented Lagrangian method of convex optimizations. Under mild conditions, we show that the proposed algorithm exhibits a linear convergence rate. Moreover, instead of verifying a computationally intractable stopping criteria, we show that the RMALM with the increasing subproblem iteration number has a global complexity $mathcalO(1/varepsilo{n}^{1+q})$ for the $varepsilon$-solution (i.e., $mathbbEleft(|x^k-x^{*}{|}^{2}ight)<varepsilon$), where $q$ is any positive number. Numerical results on synthetic and real data demonstrate that the proposed algorithm outperforms the existing algorithms.

This page was built for publication: Robins-Monro Augmented Lagrangian Method for Stochastic Convex Optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6409211)