Saved in:
Bibliographic Details
Main Authors: Borghi, Giacomo, Huang, Hui, Qiu, Jinniao
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.17373
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913443987062784
author Borghi, Giacomo
Huang, Hui
Qiu, Jinniao
author_facet Borghi, Giacomo
Huang, Hui
Qiu, Jinniao
contents We propose a zero-order optimization method for sequential min-max problems based on two populations of interacting particles. The systems are coupled so that one population aims to solve the inner maximization problem, while the other aims to solve the outer minimization problem. The dynamics are characterized by a consensus-type interaction with additional stochasticity to promote exploration of the objective landscape. Without relying on convexity or concavity assumptions, we establish theoretical convergence guarantees of the algorithm via a suitable mean-field approximation of the particle systems. Numerical experiments illustrate the validity of the proposed approach. In particular, the algorithm is able to identify a global min-max solution, in contrast to gradient-based methods, which typically converge to possibly suboptimal stationary points.
format Preprint
id arxiv_https___arxiv_org_abs_2407_17373
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A particle consensus approach to solving nonconvex-nonconcave min-max problems
Borghi, Giacomo
Huang, Hui
Qiu, Jinniao
Optimization and Control
65C35, 65K05, 90C56, 35Q90, 35Q83
We propose a zero-order optimization method for sequential min-max problems based on two populations of interacting particles. The systems are coupled so that one population aims to solve the inner maximization problem, while the other aims to solve the outer minimization problem. The dynamics are characterized by a consensus-type interaction with additional stochasticity to promote exploration of the objective landscape. Without relying on convexity or concavity assumptions, we establish theoretical convergence guarantees of the algorithm via a suitable mean-field approximation of the particle systems. Numerical experiments illustrate the validity of the proposed approach. In particular, the algorithm is able to identify a global min-max solution, in contrast to gradient-based methods, which typically converge to possibly suboptimal stationary points.
title A particle consensus approach to solving nonconvex-nonconcave min-max problems
topic Optimization and Control
65C35, 65K05, 90C56, 35Q90, 35Q83
url https://arxiv.org/abs/2407.17373