Enregistré dans:
Détails bibliographiques
Auteurs principaux: Chen, Long, Xu, Zeyi
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2510.16680
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918529275527168
author Chen, Long
Xu, Zeyi
author_facet Chen, Long
Xu, Zeyi
contents Two accelerated first-order methods, HNAG$^+$ and HNAG$^{++}$, are presented for smooth strongly convex optimization. By optimizing the coercivity constant of the HNAG flow and using a refined Lyapunov analysis, it is shown that HNAG$^+$ achieves the optimal global rate $1-2/\sqrtκ$, matching the information-theoretic lower bound for strongly convex optimization. For functions with Local Asymptotic Symmetry at the minimizer, HNAG$^{++}$ is shown to achieve the asymptotic rate $1-2\sqrt{2/κ}$, matching the best known asymptotic rate under $\mathcal C^2$ regularity, while applying to a broader local function class. Numerical experiments on linear and nonlinear examples show that the proposed methods are competitive with existing accelerated schemes.
format Preprint
id arxiv_https___arxiv_org_abs_2510_16680
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HNAG$^{++}$: An Accelerated Gradient Method with a Refined Asymptotic Rate for Strongly Convex Optimization
Chen, Long
Xu, Zeyi
Optimization and Control
Two accelerated first-order methods, HNAG$^+$ and HNAG$^{++}$, are presented for smooth strongly convex optimization. By optimizing the coercivity constant of the HNAG flow and using a refined Lyapunov analysis, it is shown that HNAG$^+$ achieves the optimal global rate $1-2/\sqrtκ$, matching the information-theoretic lower bound for strongly convex optimization. For functions with Local Asymptotic Symmetry at the minimizer, HNAG$^{++}$ is shown to achieve the asymptotic rate $1-2\sqrt{2/κ}$, matching the best known asymptotic rate under $\mathcal C^2$ regularity, while applying to a broader local function class. Numerical experiments on linear and nonlinear examples show that the proposed methods are competitive with existing accelerated schemes.
title HNAG$^{++}$: An Accelerated Gradient Method with a Refined Asymptotic Rate for Strongly Convex Optimization
topic Optimization and Control
url https://arxiv.org/abs/2510.16680