Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Long, Luo, Hao, Wei, Jingrong, Xu, Zeyi, Yao, Yuan
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2601.19038
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908790677307392
author Chen, Long
Luo, Hao
Wei, Jingrong
Xu, Zeyi
Yao, Yuan
author_facet Chen, Long
Luo, Hao
Wei, Jingrong
Xu, Zeyi
Yao, Yuan
contents Mirror descent uses the mirror function to encode geometry and constraints, improving convergence while preserving feasibility. Accelerated Mirror Descent Methods (Acc-MD) are derived from a discretization of an accelerated mirror ODE system using a variable--operator splitting framework. A geometric assumption, termed the Generalized Cauchy-Schwarz (GCS) condition, is introduced to quantify the compatibility between the objective and the mirror geometry, under which the first accelerated linear convergence for Acc-MD on a broad class of problems is established. Numerical experiments on smooth and composite optimization tasks demonstrate that Acc-MD consistently outperforms existing accelerated variants, both theoretically and empirically.
format Preprint
id arxiv_https___arxiv_org_abs_2601_19038
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Accelerated Mirror Descent Method through Variable and Operator Splitting
Chen, Long
Luo, Hao
Wei, Jingrong
Xu, Zeyi
Yao, Yuan
Optimization and Control
Mirror descent uses the mirror function to encode geometry and constraints, improving convergence while preserving feasibility. Accelerated Mirror Descent Methods (Acc-MD) are derived from a discretization of an accelerated mirror ODE system using a variable--operator splitting framework. A geometric assumption, termed the Generalized Cauchy-Schwarz (GCS) condition, is introduced to quantify the compatibility between the objective and the mirror geometry, under which the first accelerated linear convergence for Acc-MD on a broad class of problems is established. Numerical experiments on smooth and composite optimization tasks demonstrate that Acc-MD consistently outperforms existing accelerated variants, both theoretically and empirically.
title Accelerated Mirror Descent Method through Variable and Operator Splitting
topic Optimization and Control
url https://arxiv.org/abs/2601.19038