Nonasymptotic convergence analysis for the unadjusted Langevin algorithm

Abstract : In this paper, we study a method to sample from a target distribution π over Rd having a positive density with respect to the Lebesgue measure, known up to a normalisation factor. This method is based on the Euler discretization of the overdamped Langevin stochastic differential equation associated with π. For both constant and decreasing step sizes in the Euler discretization, we obtain nonasymptotic bounds for the convergence to the target distribution π in total variation distance. A particular attention is paid to the dependency on the dimension d, to demonstrate the applicability of this method in the high-dimensional setting. These bounds improve and extend the results of Dalalyan [J. R. Stat. Soc. Ser. B. Stat. Methodol. (2017) 79 651–676].
Type de document :
Article dans une revue
The Annals of Applied Probability : an official journal of the institute of mathematical statistics, The Institute of Mathematical Statistics, 2017, 27 (3), pp.1551 - 1587. 〈10.1214/16-AAP1238〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01668245
Contributeur : Eric Moulines <>
Soumis le : mardi 19 décembre 2017 - 22:36:22
Dernière modification le : jeudi 10 mai 2018 - 02:04:58

Identifiants

Citation

Alain Durmus, Eric Moulines. Nonasymptotic convergence analysis for the unadjusted Langevin algorithm. The Annals of Applied Probability : an official journal of the institute of mathematical statistics, The Institute of Mathematical Statistics, 2017, 27 (3), pp.1551 - 1587. 〈10.1214/16-AAP1238〉. 〈hal-01668245〉

Partager

Métriques

Consultations de la notice

277