Article Dans Une Revue Mathematical Programming, Series A Année : 2023

FISTA is an automatic geometrically optimized algorithm for strongly convex functions

Résumé

In this work, we are interested in the famous FISTA algorithm. We show that FISTA is an automatic geometrically optimized algorithm for functions satisfying a quadratic growth assumption. This explains why FISTA works better than the standard Forward-Backward algorithm (FB) in such a case, although FISTA is known to have a polynomial asymptotical convergence rate while FB is exponential. We provide a simple rule to tune the α parameter within the FISTA algorithm to reach an ε-solution with an optimal number of iterations. These new results highlight the efficiency of FISTA algorithms, and they rely on new non asymptotic bounds for FISTA.

Fichier principal
Vignette du fichier
automaticFISTA_Vfinale.pdf (796.89 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-03491527 , version 1 (17-12-2021)
hal-03491527 , version 2 (13-11-2024)

Licence

Identifiants

Citer

J-F Aujol, Charles Dossal, Aude Rondepierre. FISTA is an automatic geometrically optimized algorithm for strongly convex functions. Mathematical Programming, Series A, 2023, 204, pp.449-491. ⟨10.1007/s10107-023-01960-6⟩. ⟨hal-03491527v2⟩
577 Consultations
1193 Téléchargements

Altmetric

Partager

  • More