Perturbed Fenchel Duality and Primal-Dual Convergence of First-Order Methods

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Zhao, Tiantian
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908154914144256
author Zhao, Tiantian
author_facet Zhao, Tiantian
contents It has been shown that many first-order methods satisfy the perturbed Fenchel duality inequality, which yields a unified derivation of convergence. More first-order methods are discussed in this paper, e.g., dual averaging and bundle method. We show primal-dual convergence of them on convex optimization by proving the perturbed Fenchel duality property. We also propose a single-cut bundle method for saddle problem, and prove its convergence in a similar manner.
format Preprint
id arxiv_https___arxiv_org_abs_2411_09503
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Perturbed Fenchel Duality and Primal-Dual Convergence of First-Order Methods
Zhao, Tiantian
Optimization and Control
It has been shown that many first-order methods satisfy the perturbed Fenchel duality inequality, which yields a unified derivation of convergence. More first-order methods are discussed in this paper, e.g., dual averaging and bundle method. We show primal-dual convergence of them on convex optimization by proving the perturbed Fenchel duality property. We also propose a single-cut bundle method for saddle problem, and prove its convergence in a similar manner.
title Perturbed Fenchel Duality and Primal-Dual Convergence of First-Order Methods
topic Optimization and Control
url https://arxiv.org/abs/2411.09503