Automatically Marginalized MCMC in Probabilistic Programming
Machine Learning
2023-06-05 v2 Machine Learning
Abstract
Hamiltonian Monte Carlo (HMC) is a powerful algorithm to sample latent variables from Bayesian models. The advent of probabilistic programming languages (PPLs) frees users from writing inference algorithms and lets users focus on modeling. However, many models are difficult for HMC to solve directly, and often require tricks like model reparameterization. We are motivated by the fact that many of those models could be simplified by marginalization. We propose to use automatic marginalization as part of the sampling process using HMC in a graphical model extracted from a PPL, which substantially improves sampling from real-world hierarchical models.
Cite
@article{arxiv.2302.00564,
title = {Automatically Marginalized MCMC in Probabilistic Programming},
author = {Jinlin Lai and Javier Burroni and Hui Guan and Daniel Sheldon},
journal= {arXiv preprint arXiv:2302.00564},
year = {2023}
}
Comments
Accepted to the 40th International Conference on Machine Learning (ICML 2023)