An Auxiliary Variable Method for Langevin based MCMC algorithms

Yosra Marnissi 1 Emilie Chouzenoux 1 Jean-Christophe Pesquet 1 Amel Benazza-Benyahia 2
1 LIGM - Laboratoire d'Informatique Gaspard-Monge
2 URISA - unité de Recherche en Imagerie Satellitaire et ses Applications
Abstract : Markov Chain Monte Carlo sampling algorithms are efficient Bayesian tools to explore complicated posterior distributions. However, sampling in large scale problems remains a challenging task since the Markov chain is very sensitive to the dependencies between the signal samples. In this paper, we are mainly interested in Langevin based MCMC sampling algorithms that allow us to speed up the convergence by controlling the direction of sampling and/or exploiting the correlation structure of the target signal. However, these techniques may sometimes fail to explore efficiently the target space because of poor mixing properties of the chain or the high cost of each iteration. By adding some auxiliary variables, we show that the resulting conditional distribution of the target signal is much simpler to explore by using these algorithms. Experiments performed in the context of multicomponent image restoration illustrate that the proposed approach can achieve substantial performance improvement compared with standard algorithms.
Type de document :
Communication dans un congrès
IEEE Workshop on Statistical Signal Processing (SSP 2016), Jun 2016, Palma de Mallorque, Spain. Proceedings of the IEEE Workshop on Statistical Signal Processing (SSP 2016), pp.297--301, 2016, Proceedings of the IEEE Workshop on Statistical Signal Processing (SSP 2016)
Liste complète des métadonnées
https://hal.archives-ouvertes.fr/hal-01386560
Contributeur : Emilie Chouzenoux <>
Soumis le : lundi 24 octobre 2016 - 11:56:50
Dernière modification le : mercredi 11 avril 2018 - 12:12:03

Identifiants

  • HAL Id : hal-01386560, version 1

Collections

UPEC-UPEM | LIGM_SIGNAL | LIGM | ENPC | PARISTECH

Citation

Yosra Marnissi, Emilie Chouzenoux, Jean-Christophe Pesquet, Amel Benazza-Benyahia. An Auxiliary Variable Method for Langevin based MCMC algorithms. IEEE Workshop on Statistical Signal Processing (SSP 2016), Jun 2016, Palma de Mallorque, Spain. Proceedings of the IEEE Workshop on Statistical Signal Processing (SSP 2016), pp.297--301, 2016, Proceedings of the IEEE Workshop on Statistical Signal Processing (SSP 2016). 〈hal-01386560〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

95

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD