Variance Reduction Using Nonreversible Langevin Samplers - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Journal of Statistical Physics Année : 2016

Variance Reduction Using Nonreversible Langevin Samplers

Résumé

A standard approach to computing expectations with respect to a given target measure is to introduce an overdamped Langevin equation which is reversible with respect to the target distribution, and to approximate the expectation by a time-averaging estimator. As has been noted in recent papers, introducing an appropriately chosen nonreversible component to the dynamics is beneficial, both in terms of reducing the asymptotic variance and of speeding up convergence to the target distribution. In this paper we present a detailed study of the dependence of the asymptotic variance on the deviation from reversibility. Our theoretical findings are supported by numerical simulations.

Dates et versions

hal-01164466 , version 1 (17-06-2015)

Identifiants

Citer

Andrew B. Duncan, Tony Lelièvre, Grigorios A. Pavliotis. Variance Reduction Using Nonreversible Langevin Samplers. Journal of Statistical Physics, 2016, 163 (3), pp.457 - 491. ⟨10.1007/s10955-016-1491-2⟩. ⟨hal-01164466⟩
127 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More