Component-wise approximate Bayesian computation via Gibbs-like steps

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Clarté, Grégoire, Robert, Christian P., Ryder, Robin, Stoehr, Julien
Format: Preprint
Publié: 2019
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912883057623040
author Clarté, Grégoire
Robert, Christian P.
Ryder, Robin
Stoehr, Julien
author_facet Clarté, Grégoire
Robert, Christian P.
Ryder, Robin
Stoehr, Julien
contents Approximate Bayesian computation methods are useful for generative models with intractable likelihoods. These methods are however sensitive to the dimension of the parameter space, requiring exponentially increasing resources as this dimension grows. To tackle this difficulty, we explore a Gibbs version of the ABC approach that runs component-wise approximate Bayesian computation steps aimed at the corresponding conditional posterior distributions, and based on summary statistics of reduced dimensions. While lacking the standard justifications for the Gibbs sampler, the resulting Markov chain is shown to converge in distribution under some partial independence conditions. The associated stationary distribution can further be shown to be close to the true posterior distribution and some hierarchical versions of the proposed mechanism enjoy a closed form limiting distribution. Experiments also demonstrate the gain in efficiency brought by the Gibbs version over the standard solution.
format Preprint
id arxiv_https___arxiv_org_abs_1905_13599
institution arXiv
publishDate 2019
record_format arxiv
spellingShingle Component-wise approximate Bayesian computation via Gibbs-like steps
Clarté, Grégoire
Robert, Christian P.
Ryder, Robin
Stoehr, Julien
Computation
Methodology
Approximate Bayesian computation methods are useful for generative models with intractable likelihoods. These methods are however sensitive to the dimension of the parameter space, requiring exponentially increasing resources as this dimension grows. To tackle this difficulty, we explore a Gibbs version of the ABC approach that runs component-wise approximate Bayesian computation steps aimed at the corresponding conditional posterior distributions, and based on summary statistics of reduced dimensions. While lacking the standard justifications for the Gibbs sampler, the resulting Markov chain is shown to converge in distribution under some partial independence conditions. The associated stationary distribution can further be shown to be close to the true posterior distribution and some hierarchical versions of the proposed mechanism enjoy a closed form limiting distribution. Experiments also demonstrate the gain in efficiency brought by the Gibbs version over the standard solution.
title Component-wise approximate Bayesian computation via Gibbs-like steps
topic Computation
Methodology
url https://arxiv.org/abs/1905.13599