Gradient scan Gibbs sampler: An efficient high-dimensional sampler application in inverse problems

Abstract : The paper deals with Gibbs samplers that include high-dimensional conditional Gaussian distributions. It proposes an efficient algorithm that only requires a scalar Gaussian sampling. The algorithm relies on a random excursion along a random direction. It is proved to converge, i.e. the drawn samples are asymptotically under the target distribution. Our original motivation is in unsupervised inverse problems related to general linear observation models and their solution in a hierarchical Bayesian framework implemented through sampling algorithms. The paper provides an illustration focused on 2-D simulations and on the super-resolution problem.
Type de document :
Communication dans un congrès
40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2015), Apr 2015, Brisbane, Australia. ICASSP 2015, 2015, ICASSP 2015. 〈10.1109/ICASSP.2015.7178739〉
Liste complète des métadonnées

Littérature citée [26 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01225866
Contributeur : François Orieux <>
Soumis le : vendredi 6 novembre 2015 - 21:28:01
Dernière modification le : vendredi 21 décembre 2018 - 11:06:09
Document(s) archivé(s) le : lundi 8 février 2016 - 10:54:37

Fichier

Orieux15 - Gradient scan gibbs...
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

François Orieux, O Féron, Jean-François Giovannelli. Gradient scan Gibbs sampler: An efficient high-dimensional sampler application in inverse problems. 40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2015), Apr 2015, Brisbane, Australia. ICASSP 2015, 2015, ICASSP 2015. 〈10.1109/ICASSP.2015.7178739〉. 〈hal-01225866〉

Partager

Métriques

Consultations de la notice

366

Téléchargements de fichiers

185