Mixed Hamiltonian Monte Carlo for Mixed Discrete and Continuous Variables

Abstract

Hamiltonian Monte Carlo (HMC) has emerged as a powerful Markov Chain Monte Carlo (MCMC) method to sample from complex continuous distributions. However, a fundamental limitation of HMC is that it can not be applied to distributions with mixed discrete and continuous variables. In this paper, we propose mixed HMC (M-HMC) as a general framework to address this limitation. M-HMC is a novel family of MCMC algorithms that evolves the discrete and continuous variables in tandem, allowing more frequent updates of discrete variables while maintaining HMC's ability to suppress random-walk behavior. We establish M-HMC's theoretical properties, and present an efficient implementation with Laplace momentum that introduces minimal overhead compared to existing HMC methods. The superior performances of M-HMC over existing methods are demonstrated with numerical experiments on Gaussian mixture models (GMMs), variable selection in Bayesian logistic regression (BLR), and correlated topic models (CTMs).

Results with different discrete proposals

References in corpus (6)

• Pyro: Deep Universal Probabilistic Programming
• A Conceptual Introduction to Hamiltonian Monte Carlo
• Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro
• Discontinuous Hamiltonian Monte Carlo for discrete parameters and discontinuous likelihoods
• Probabilistic Path Hamiltonian Monte Carlo
• LF-PPL: A Low-Level First Order Probabilistic Programming Language for Non-Differentiable Models