Lipschitz-free Projected Subgradient Method with Time-varying Step-size

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Xia, Yong, Zhang, Yanhao, Zhu, Zhihan
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916949373485056
author Xia, Yong
Zhang, Yanhao
Zhu, Zhihan
author_facet Xia, Yong
Zhang, Yanhao
Zhu, Zhihan
contents We introduce a novel family of time-varying step-sizes for the classical projected subgradient method, offering optimal ergodic convergence. Importantly, this approach does not depend on the Lipschitz assumption of the objective function, thereby broadening the convergence result of projected subgradient method to non-Lipschitz case.
format Preprint
id arxiv_https___arxiv_org_abs_2410_22336
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Lipschitz-free Projected Subgradient Method with Time-varying Step-size
Xia, Yong
Zhang, Yanhao
Zhu, Zhihan
Optimization and Control
90C25, 90C30
We introduce a novel family of time-varying step-sizes for the classical projected subgradient method, offering optimal ergodic convergence. Importantly, this approach does not depend on the Lipschitz assumption of the objective function, thereby broadening the convergence result of projected subgradient method to non-Lipschitz case.
title Lipschitz-free Projected Subgradient Method with Time-varying Step-size
topic Optimization and Control
90C25, 90C30
url https://arxiv.org/abs/2410.22336