Saved in:
Bibliographic Details
Main Authors: Heng, Qiang, Liu, Xiaoqian, Chi, Eric C.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.14269
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916622498791424
author Heng, Qiang
Liu, Xiaoqian
Chi, Eric C.
author_facet Heng, Qiang
Liu, Xiaoqian
Chi, Eric C.
contents Convex-nonconvex (CNC) regularization is a novel paradigm that employs a nonconvex penalty function while maintaining the convexity of the entire objective function. It has been successfully applied to problems in signal processing, statistics, and machine learning. Despite its wide application, the computation of CNC regularized problems remains challenging and under-investigated. To fill the gap, we study several operator splitting methods and their Anderson accelerated counterparts for solving least squares problems with CNC regularization. We establish the global convergence of the proposed algorithm to an optimal point and demonstrate its practical speed-ups in various applications.
format Preprint
id arxiv_https___arxiv_org_abs_2502_14269
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Anderson Accelerated Operator Splitting Methods for Convex-nonconvex Regularized Problems
Heng, Qiang
Liu, Xiaoqian
Chi, Eric C.
Optimization and Control
Computation
Convex-nonconvex (CNC) regularization is a novel paradigm that employs a nonconvex penalty function while maintaining the convexity of the entire objective function. It has been successfully applied to problems in signal processing, statistics, and machine learning. Despite its wide application, the computation of CNC regularized problems remains challenging and under-investigated. To fill the gap, we study several operator splitting methods and their Anderson accelerated counterparts for solving least squares problems with CNC regularization. We establish the global convergence of the proposed algorithm to an optimal point and demonstrate its practical speed-ups in various applications.
title Anderson Accelerated Operator Splitting Methods for Convex-nonconvex Regularized Problems
topic Optimization and Control
Computation
url https://arxiv.org/abs/2502.14269