Uncertainty Quantification in Bayesian Clustering

TL;DR

This paper introduces a practical post-processing method for summarizing and interpreting uncertainty in Bayesian clustering, addressing the challenge of complex posterior distributions over partitions.

Contribution

It proposes a new, efficient procedure to generate credible sets and measures of uncertainty for Bayesian clustering, avoiding label-switching issues.

Findings

Provides a fast, user-friendly credible set construction

Introduces new measures for clustering uncertainty

Demonstrates effectiveness through multiple applications

Abstract

Bayesian clustering methods have the widely touted advantage of providing a probabilistic characterization of uncertainty in clustering through the posterior distribution. An amazing variety of priors and likelihoods have been proposed for clustering in a broad array of settings. There is also a rich literature on Markov chain Monte Carlo (MCMC) algorithms for sampling from posterior clustering distributions. However, there is relatively little work on summarizing the posterior uncertainty. The complexity of the partition space corresponding to different clusterings makes this problem challenging. We propose a post-processing procedure for any Bayesian clustering model with posterior samples that generates a credible set that is easy to use, fast to compute, and intuitive to interpret. We also provide new measures of clustering uncertainty and show how to compute cluster-specific parameter estimates and credible regions that accumulate a desired posterior probability without having to condition on a partition estimate or employ label-switching techniques. We illustrate our procedure through several applications.