Entropy-based adaptive Hamiltonian Monte Carlo

TL;DR

This paper introduces an entropy-based adaptive method for Hamiltonian Monte Carlo that optimizes the mass matrix to improve sampling efficiency and acceptance rates by leveraging gradient information.

Contribution

It proposes a novel gradient-based adaptation strategy for the mass matrix in HMC that maximizes proposal entropy, enhancing sampling performance.

Findings

Outperforms standard HMC schemes in experiments

Adapts to the geometry of the target distribution

Provides better control over integration time

Abstract

Hamiltonian Monte Carlo (HMC) is a popular Markov Chain Monte Carlo (MCMC) algorithm to sample from an unnormalized probability distribution. A leapfrog integrator is commonly used to implement HMC in practice, but its performance can be sensitive to the choice of mass matrix used therein. We develop a gradient-based algorithm that allows for the adaptation of the mass matrix by encouraging the leapfrog integrator to have high acceptance rates while also exploring all dimensions jointly. In contrast to previous work that adapt the hyperparameters of HMC using some form of expected squared jumping distance, the adaptation strategy suggested here aims to increase sampling efficiency by maximizing an approximation of the proposal entropy. We illustrate that using multiple gradients in the HMC proposal can be beneficial compared to a single gradient-step in Metropolis-adjusted Langevin proposals. Empirical evidence suggests that the adaptation method can outperform different versions of HMC schemes by adjusting the mass matrix to the geometry of the target distribution and by providing some control on the integration time.