An ABC interpretation of the multiple auxiliary variable method

TL;DR

This paper demonstrates that the auxiliary variable method for Markov random fields can be interpreted as an approximate Bayesian computation approach for likelihood estimation, offering a new perspective on its theoretical foundation.

Contribution

It provides a novel interpretation of the auxiliary variable method as an ABC technique, bridging two statistical inference frameworks.

Findings

Reveals the connection between auxiliary variable methods and ABC.

Offers a new theoretical perspective on likelihood estimation.

Enhances understanding of Markov random fields inference.

Abstract

We show that the auxiliary variable method (M{\o}ller et al., 2006; Murray et al., 2006) for inference of Markov random fields can be viewed as an approximate Bayesian computation method for likelihood estimation.