EUVIMEDEUVIMED entdecken
Crossref

Complexity-Optimal and Parameter-Free First-Order Methods for Finding Stationary Points of Composite Optimization Problems

Weiwei Kong

SIAM Journal on Optimization · 30.09.2024 · 34 (3) · 3005-3032

DOI: 10.1137/22m1498826 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SIAM Journal on Optimization, online erschienen am 04.09.2024. Verfasst von Weiwei Kong. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 48 Literaturangaben. Gefördert von: National Nuclear Security Administration.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper develops and analyzes an accelerated proximal descent method for finding stationary points of nonconvex composite optimization problems. The objective function is of the form [Formula: see text], where [Formula: see text] is a proper closed convex function, [Formula: see text] is a differentiable function on the domain of [Formula: see text], and [Formula: see text] is Lipschitz continuous on the domain of [Formula: see text]. The main advantage of this method is that it is “parameter-free” in the sense that it does not require knowledge of the Lipschitz constant of [Formula: see text] or of any global topological properties of [Formula: see text]. It is shown that the proposed method can obtain an [Formula: see text]-approximate stationary point with iteration complexity bounds that are optimal, up to logarithmic terms over [Formula: see text], in both the convex and nonconvex settings. Some discussion is also given about how the proposed method can be leveraged in other existing optimization frameworks, such as min-max smoothing and penalty frameworks for constrained programming, to create more specialized parameter-free methods. Finally, numerical experiments are presented to support the practical viability of the method.

Förderung

  • National Nuclear Security Administration

Zitieren

APA

Kong, W. (2024). Complexity-Optimal and Parameter-Free First-Order Methods for Finding Stationary Points of Composite Optimization Problems. SIAM Journal on Optimization, 34(3), 3005-3032. https://doi.org/10.1137/22m1498826

Vancouver

Kong W. Complexity-Optimal and Parameter-Free First-Order Methods for Finding Stationary Points of Composite Optimization Problems. SIAM Journal on Optimization. 2024;34(3):3005-3032. doi:10.1137/22m1498826

BibTeX
@article{Kong2024,
  author = {Kong, Weiwei},
  title = {Complexity-Optimal and Parameter-Free First-Order Methods for Finding Stationary Points of Composite Optimization Problems},
  journal = {SIAM Journal on Optimization},
  year = {2024},
  volume = {34},
  number = {3},
  pages = {3005--3032},
  doi = {10.1137/22m1498826}
}
Kostenlos starten →