Avoiding Strict Saddle Points of Nonconvex Regularized Problems
SIAM Journal on Optimization · 30.06.2026 · 36 (2) · 938-967
DOI: 10.1137/24m1707687 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Optimization, online erschienen am 29.05.2026. Verfasst von Luwei Bai und 3 weiteren Autor:innen. Die Arbeit nennt 56 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Shenzhen Medical Research Fund, NSFC.
Abstract (Originalsprache)
This paper considers a class of nonconvex and nonsmooth sparse optimization problems, encompassing most existing nonconvex sparsity-inducing terms. We show that their second-order optimality conditions depend only on the nonzeros of the stationary points. We propose two damped iteratively reweighted algorithms, which are the iteratively reweighted [Formula: see text] algorithm (DIRL[Formula: see text]) and the iteratively reweighted [Formula: see text] (DIRL[Formula: see text]) algorithm, to solve these problems. For DIRL[Formula: see text], we show that the reweighted [Formula: see text] subproblem has the support identification property so that DIRL[Formula: see text] locally reverts to a gradient descent algorithm around a stationary point. For DIRL[Formula: see text], we show that the solution map of the reweighted [Formula: see text] subproblem is differentiable and Lipschitz continuous everywhere. Therefore, the solution maps of DIRL[Formula: see text] and DIRL[Formula: see text] and their inverses are Lipschitz continuous, and the strict saddle points are their unstable fixed points. By applying the stable manifold theorem, these algorithms starting from almost every initial point are shown to converge to local minima when the strict saddle point property is assumed.
Förderung
- National Natural Science Foundation of China
- Shenzhen Medical Research Fund
- NSFC
- Educational Commission of Guangdong Province
- Shenzhen Science and Technology Program
- Shenzhen University
- Research Grants Council of Hong Kong
Zitieren
APA
Bai, L., Hu, Y., Wang, H., & Yang, X. (2026). Avoiding Strict Saddle Points of Nonconvex Regularized Problems. SIAM Journal on Optimization, 36(2), 938-967. https://doi.org/10.1137/24m1707687
Vancouver
Bai L, Hu Y, Wang H, Yang X. Avoiding Strict Saddle Points of Nonconvex Regularized Problems. SIAM Journal on Optimization. 2026;36(2):938-967. doi:10.1137/24m1707687
BibTeX
@article{Bai2026,
author = {Bai, Luwei and Hu, Yaohua and Wang, Hao and Yang, Xiaoqi},
title = {Avoiding Strict Saddle Points of Nonconvex Regularized Problems},
journal = {SIAM Journal on Optimization},
year = {2026},
volume = {36},
number = {2},
pages = {938--967},
doi = {10.1137/24m1707687}
}
EUVIMED