On the accept-reject mechanism for Metropolis-Hastings algorithms

TL;DR

This paper introduces a unified framework for acceptance ratios in Metropolis-Hastings algorithms, enabling the development of new kernels and revealing connections between various sampling methods, including Hamiltonian dynamics.

Contribution

It presents a general theoretical framework that unifies different Metropolis-Hastings kernels, including Hamiltonian-based and diffusion-type methods, facilitating the creation of novel algorithms.

Findings

Derived new classes of Metropolis-Hastings kernels.

Unified existing sampling methods under a common theoretical framework.

Established equivalence with earlier state space results.

Abstract

This work develops a powerful and versatile framework for determining acceptance ratios in Metropolis-Hastings type Markov kernels widely used in statistical sampling problems. Our approach allows us to derive new classes of kernels which unify random walk or diffusion-type sampling methods with more complicated "extended phase space" algorithms based around ideas from Hamiltonian dynamics. Our starting point is an abstract result developed in the generality of measurable state spaces that addresses proposal kernels that possess a certain involution structure. Note that, while this underlying proposal structure suggests a scope which includes Hamiltonian-type kernels, we demonstrate that our abstract result is, in an appropriate sense, equivalent to an earlier general state space setting developed in [Tierney, Annals of Applied Probability, 1998] where the connection to Hamiltonian methods was more obscure. Altogether, the theoretical unity and reach of our main result provides a basis for deriving novel sampling algorithms while laying bare important relationships between existing methods.