Linear Convergence of a Unified Primal--Dual Algorithm for Convex--Concave Saddle Point Problems with Quadratic Growth

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Melcher, Cody, Jalilzadeh, Afrooz, Hamedani, Erfan Yazdandoost
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911208776400896
author Melcher, Cody
Jalilzadeh, Afrooz
Hamedani, Erfan Yazdandoost
author_facet Melcher, Cody
Jalilzadeh, Afrooz
Hamedani, Erfan Yazdandoost
contents In this paper, we study saddle point (SP) problems, focusing on convex-concave optimization involving functions that satisfy either two-sided quadratic functional growth (QFG) or two-sided quadratic gradient growth (QGG)--novel conditions tailored specifically for SP problems as extensions of quadratic growth conditions in minimization. These conditions relax the traditional requirement of strong convexity-strong concavity, thereby encompassing a broader class of problems. We propose a generalized accelerated primal-dual (GAPD) algorithm to solve SP problems with non-bilinear objective functions, unifying and extending existing methods. We prove that our method achieves a linear convergence rate under these relaxed conditions. Additionally, we provide examples of structured SP problems that satisfy either two-sided QFG or QGG, demonstrating the practical applicability and relevance of our approach.
format Preprint
id arxiv_https___arxiv_org_abs_2510_11990
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Linear Convergence of a Unified Primal--Dual Algorithm for Convex--Concave Saddle Point Problems with Quadratic Growth
Melcher, Cody
Jalilzadeh, Afrooz
Hamedani, Erfan Yazdandoost
Optimization and Control
In this paper, we study saddle point (SP) problems, focusing on convex-concave optimization involving functions that satisfy either two-sided quadratic functional growth (QFG) or two-sided quadratic gradient growth (QGG)--novel conditions tailored specifically for SP problems as extensions of quadratic growth conditions in minimization. These conditions relax the traditional requirement of strong convexity-strong concavity, thereby encompassing a broader class of problems. We propose a generalized accelerated primal-dual (GAPD) algorithm to solve SP problems with non-bilinear objective functions, unifying and extending existing methods. We prove that our method achieves a linear convergence rate under these relaxed conditions. Additionally, we provide examples of structured SP problems that satisfy either two-sided QFG or QGG, demonstrating the practical applicability and relevance of our approach.
title Linear Convergence of a Unified Primal--Dual Algorithm for Convex--Concave Saddle Point Problems with Quadratic Growth
topic Optimization and Control
url https://arxiv.org/abs/2510.11990