Integral control of the proximal gradient method for unbiased sparse optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cerone, V., Fosson, S. M., Re, A., Regruto, D.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916694397550592
author Cerone, V.
Fosson, S. M.
Re, A.
Regruto, D.
author_facet Cerone, V.
Fosson, S. M.
Re, A.
Regruto, D.
contents Proximal gradient methods are popular in sparse optimization as they are straightforward to implement. Nevertheless, they achieve biased solutions, requiring many iterations to converge. This work addresses these issues through a suitable feedback control of the algorithm's hyperparameter. Specifically, by designing an integral control that does not substantially impact the computational complexity, we can reach an unbiased solution in a reasonable number of iterations. In the paper, we develop and analyze the convergence of the proposed approach for strongly-convex problems. Moreover, numerical simulations validate and extend the theoretical results to the non-strongly convex framework.
format Preprint
id arxiv_https___arxiv_org_abs_2504_12814
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Integral control of the proximal gradient method for unbiased sparse optimization
Cerone, V.
Fosson, S. M.
Re, A.
Regruto, D.
Optimization and Control
Systems and Control
Proximal gradient methods are popular in sparse optimization as they are straightforward to implement. Nevertheless, they achieve biased solutions, requiring many iterations to converge. This work addresses these issues through a suitable feedback control of the algorithm's hyperparameter. Specifically, by designing an integral control that does not substantially impact the computational complexity, we can reach an unbiased solution in a reasonable number of iterations. In the paper, we develop and analyze the convergence of the proposed approach for strongly-convex problems. Moreover, numerical simulations validate and extend the theoretical results to the non-strongly convex framework.
title Integral control of the proximal gradient method for unbiased sparse optimization
topic Optimization and Control
Systems and Control
url https://arxiv.org/abs/2504.12814