Online, Informative MCMC Thinning with Kernelized Stein Discrepancy
Abstract
A fundamental challenge in Bayesian inference is efficient representation of a target distribution. Many non-parametric approaches do so by sampling a large number of points using variants of Markov Chain Monte Carlo (MCMC). We propose an MCMC variant that retains only those posterior samples which exceed a KSD threshold, which we call KSD Thinning. We establish the convergence and complexity tradeoffs for several settings of KSD Thinning as a function of the KSD threshold parameter, sample size, and other problem parameters. Finally, we provide experimental comparisons against other online nonparametric Bayesian methods that generate low-complexity posterior representations, and observe superior consistency/complexity tradeoffs. Code is available at github.com/colehawkins/KSD-Thinning.
Cite
@article{arxiv.2201.07130,
title = {Online, Informative MCMC Thinning with Kernelized Stein Discrepancy},
author = {Cole Hawkins and Alec Koppel and Zheng Zhang},
journal= {arXiv preprint arXiv:2201.07130},
year = {2022}
}
Comments
New version corrects error in notation regarding dictionary size growth. Little o to Omega