Extra Chance Generalized Hybrid Monte Carlo

TL;DR

This paper introduces Extra Chance Generalized Hybrid Monte Carlo, a method that reduces rejections in HMC algorithms by performing additional proposals, leading to higher quality samples.

Contribution

It proposes a novel rejection avoidance technique for HMC algorithms that improves sample quality through extra proposal attempts.

Findings

Extra work per sample improves sample quality.

The method effectively reduces rejection rates.

Samples are more representative of the target distribution.

Abstract

We study a method, Extra Chance Generalized Hybrid Monte Carlo, to avoid rejections in the Hybrid Monte Carlo method and related algorithms. In the spirit of delayed rejection, whenever a rejection would occur, extra work is done to find a fresh proposal that, hopefully, may be accepted. We present experiments that clearly indicate that the additional work per sample carried out in the extra chance approach clearly pays in terms of the quality of the samples generated.