Bayesian cluster analysis: Point estimation and credible balls

Abstract

Clustering is widely studied in statistics and machine learning, with applications in a variety of fields. As opposed to classical algorithms which return a single clustering solution, Bayesian nonparametric models provide a posterior over the entire space of partitions, allowing one to assess statistical properties, such as uncertainty on the number of clusters. However, an important problem is how to summarize the posterior; the huge dimension of partition space and difficulties in visualizing it add to this problem. In a Bayesian analysis, the posterior of a real-valued parameter of interest is often summarized by reporting a point estimate such as the posterior mean along with 95% credible intervals to characterize uncertainty. In this paper, we extend these ideas to develop appropriate point estimates and credible sets to summarize the posterior of clustering structure based on decision and information theoretic techniques.

References in corpus (7)

• Natural Scales in Geographical Patterns
• MCMC for Normalized Random Measure Mixture Models
• A simple example of Dirichlet process mixture inconsistency for the number of components
• MAD-Bayes: MAP-based Asymptotic Derivations from Bayes
• Inconsistency of Pitman-Yor process mixtures for the number of components
• Simple approximate MAP Inference for Dirichlet processes
• A hybrid sampler for Poisson-Kingman mixture models

Cited by in corpus (28)

• Flexible clustering via hidden hierarchical Dirichlet priors
• Bayesian Distance Clustering
• Escaping the curse of dimensionality in Bayesian model based clustering
• Spying on the prior of the number of data clusters and the partition distribution in Bayesian cluster analysis
• Latent Simplex Position Model: High Dimensional Multi-view Clustering with Uncertainty Quantification
• Cohesion and Repulsion in Bayesian Distance Clustering
• Optimal Clustering under Uncertainty
• Cross-Study Replicability in Cluster Analysis
• A Bayesian Nonparametric Model for Predicting Pregnancy Outcomes Using Longitudinal Profiles
• A Two-Stage Bayesian Semiparametric Model for Novelty Detection with Robust Prior Information
• Distributed Bayesian clustering using finite mixture of mixtures
• Entropy regularization in probabilistic clustering
• Optimal Bayesian clustering using non-negative matrix factorization
• Root and community inference on the latent growth process of a network
• Bayesian Nonparametric Models for Multiple Raters: a General Statistical Framework
• General Bayesian inference schemes in infinite mixture models
• Generalized infinite factorization models
• Model based clustering of multinomial count data
• A Bayesian Nonparametric Stochastic Block Model for Directed Acyclic Graphs
• Graph-Aligned Random Partition Model (GARP)
• Consensus Monte Carlo for Random Subsets using Shared Anchors
• A Common Atom Model for the Bayesian Nonparametric Analysis of Nested Data
• Parameter clustering in Bayesian functional PCA of fMRI data
• The modeling of multiple animals that share behavioral features
• A Bayesian Model for Co-clustering Ordinal Data with Informative Missing Entries
• Heterogeneous Effects in the Built Environment
• BNPdensity: Bayesian nonparametric mixture modeling in R
• Bayesian Fusion Estimation via t-Shrinkage