Evidence and Bayes factor estimation for Gibbs random fields

TL;DR

This paper develops a new method for estimating evidence and Bayes factors in Gibbs random fields, improving Bayesian inference for these complex models with intractable likelihoods.

Contribution

It introduces an approach for estimating evidence and Bayes factors specifically tailored for Gibbs random fields, advancing computational Bayesian methods.

Findings

The proposed method yields good performance in estimating evidence.

It effectively handles intractable likelihoods in Gibbs random fields.

The approach improves Bayesian inference accuracy for complex models.

Abstract

Gibbs random fields play an important role in statistics. However they are complicated to work with due to an intractability of the likelihood function and there has been much work devoted to finding computational algorithms to allow Bayesian inference to be conducted for such so-called doubly intractable distributions. This paper extends this work and addresses the issue of estimating the evidence and Bayes factor for such models. The approach which we develop is shown to yield good performance.