A systematic approach to Lyapunov analyses of continuous-time models in convex optimization - ENS - École normale supérieure Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2022

A systematic approach to Lyapunov analyses of continuous-time models in convex optimization

Résumé

First-order methods are often analyzed via their continuous-time models, where their worst-case convergence properties are usually approached via Lyapunov functions. In this work, we provide a systematic and principled approach to find and verify Lyapunov functions for classes of ordinary and stochastic differential equations. More precisely, we extend the performance estimation framework, originally proposed by Drori and Teboulle [10], to continuous-time models. We retrieve convergence results comparable to those of discrete methods using fewer assumptions and convexity inequalities, and provide new results for stochastic accelerated gradient flows.
Fichier principal
Vignette du fichier
main.pdf (480.76 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03677528 , version 1 (24-05-2022)
hal-03677528 , version 2 (25-05-2022)
hal-03677528 , version 3 (07-03-2024)

Identifiants

Citer

Céline Moucer, Adrien Taylor, Francis Bach. A systematic approach to Lyapunov analyses of continuous-time models in convex optimization. 2022. ⟨hal-03677528v2⟩
115 Consultations
174 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More