Simple linesearch-free first-order methods for nonconvex optimization

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Yagishita, Shotaro, Ito, Masaru
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912591530426368
author Yagishita, Shotaro
Ito, Masaru
author_facet Yagishita, Shotaro
Ito, Masaru
contents This paper presents an auto-conditioned proximal gradient method for nonconvex optimization. The method determines the stepsize using an estimation of local curvature and does not require any prior knowledge of problem parameters and any linesearch procedures. Its convergence analysis is carried out in a simple manner without assuming the convexity, unlike previous studies. We also provide convergence analysis in the presence of the Kurdyka--Łojasiewicz property, adaptivity to the weak smoothness, and the extension to the Bregman proximal gradient method. Furthermore, the auto-conditioned stepsize strategy is also applied to the conditional gradient (Frank--Wolfe) method and the Riemannian gradient method.
format Preprint
id arxiv_https___arxiv_org_abs_2509_14670
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Simple linesearch-free first-order methods for nonconvex optimization
Yagishita, Shotaro
Ito, Masaru
Optimization and Control
This paper presents an auto-conditioned proximal gradient method for nonconvex optimization. The method determines the stepsize using an estimation of local curvature and does not require any prior knowledge of problem parameters and any linesearch procedures. Its convergence analysis is carried out in a simple manner without assuming the convexity, unlike previous studies. We also provide convergence analysis in the presence of the Kurdyka--Łojasiewicz property, adaptivity to the weak smoothness, and the extension to the Bregman proximal gradient method. Furthermore, the auto-conditioned stepsize strategy is also applied to the conditional gradient (Frank--Wolfe) method and the Riemannian gradient method.
title Simple linesearch-free first-order methods for nonconvex optimization
topic Optimization and Control
url https://arxiv.org/abs/2509.14670