Non-Convex Global Optimization as an Optimal Stabilization Problem: Dynamical Properties

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Huang, Yuyang, Kalise, Dante, Kouhkouh, Hicham
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915616645971968
author Huang, Yuyang
Kalise, Dante
Kouhkouh, Hicham
author_facet Huang, Yuyang
Kalise, Dante
Kouhkouh, Hicham
contents We study global optimization of non-convex functions through optimal control theory. Our main result establishes that (quasi-)optimal trajectories of a discounted control problem converge globally and practically asymptotically to the set of global minimizers. Specifically, for any tolerance $η> 0$, there exist parameters $λ$ (discount rate) and $t$ (time horizon) such that trajectories remain within an $η$-neighborhood of the global minimizers after some finite time $τ$. This convergence is achieved directly, without solving ergodic Hamilton-Jacobi-Bellman equations. We prove parallel results for three problem formulations: evolutive discounted, stationary discounted, and evolutive non-discounted cases. The analysis relies on occupation measures to quantify the fraction of time trajectories spend away from the minimizer set, establishing both reachability and stability properties.
format Preprint
id arxiv_https___arxiv_org_abs_2511_10815
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Non-Convex Global Optimization as an Optimal Stabilization Problem: Dynamical Properties
Huang, Yuyang
Kalise, Dante
Kouhkouh, Hicham
Optimization and Control
37N35, 90C26, 49L12, 35Q93
We study global optimization of non-convex functions through optimal control theory. Our main result establishes that (quasi-)optimal trajectories of a discounted control problem converge globally and practically asymptotically to the set of global minimizers. Specifically, for any tolerance $η> 0$, there exist parameters $λ$ (discount rate) and $t$ (time horizon) such that trajectories remain within an $η$-neighborhood of the global minimizers after some finite time $τ$. This convergence is achieved directly, without solving ergodic Hamilton-Jacobi-Bellman equations. We prove parallel results for three problem formulations: evolutive discounted, stationary discounted, and evolutive non-discounted cases. The analysis relies on occupation measures to quantify the fraction of time trajectories spend away from the minimizer set, establishing both reachability and stability properties.
title Non-Convex Global Optimization as an Optimal Stabilization Problem: Dynamical Properties
topic Optimization and Control
37N35, 90C26, 49L12, 35Q93
url https://arxiv.org/abs/2511.10815