Stochastic Halpern Iteration with Variance Reduction for Stochastic Monotone Inclusions

arXiv2203.09436MaRDI QIDQ6393993

Author name not available (Why is that?)

Publication date: 17 March 2022

Abstract: We study stochastic monotone inclusion problems, which widely appear in machine learning applications, including robust regression and adversarial learning. We propose novel variants of stochastic Halpern iteration with recursive variance reduction. In the cocoercive -- and more generally Lipschitz-monotone -- setup, our algorithm attains $epsilon$ norm of the operator with $mathcalO(frac1epsilo{n}^3)$ stochastic operator evaluations, which significantly improves over state of the art $mathcalO(frac1epsilo{n}^4)$ stochastic operator evaluations required for existing monotone inclusion solvers applied to the same problem classes. We further show how to couple one of the proposed variants of stochastic Halpern iteration with a scheduled restart scheme to solve stochastic monotone inclusion problems with $mathcalO(fraclog(1/epsilon)epsilo{n}^2)$ stochastic operator evaluations under additional sharpness or strong monotonicity assumptions.

Has companion code repository: https://github.com/zephyr-cai/halpern

This page was built for publication: Stochastic Halpern Iteration with Variance Reduction for Stochastic Monotone Inclusions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6393993)