Convergence of latent mixing measures in finite and infinite mixture models

Abstract

This paper studies convergence behavior of latent mixing measures that arise in finite and infinite mixture models, using transportation distances (i.e., Wasserstein metrics). The relationship between Wasserstein distances on the space of mixing measures and f-divergence functionals such as Hellinger and Kullback-Leibler distances on the space of mixture distributions is investigated in detail using various identifiability conditions. Convergence in Wasserstein metrics for discrete measures implies convergence of individual atoms that provide support for the measures, thereby providing a natural interpretation of convergence of clusters in clustering applications where mixture models are typically employed. Convergence rates of posterior distributions for latent mixing measures are established, for both finite mixtures of multivariate distributions and infinite mixtures based on the Dirichlet process.

Published in at http://dx.doi.org/10.1214/12-AOS1065 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (3)

• Convergence rates of posterior distributions for noniid observations
• Posterior convergence rates of Dirichlet mixtures at smooth densities
• On rates of convergence for posterior distributions in infinite-dimensional models

Cited by in corpus (13)

• Clustering consistency with Dirichlet process mixtures
• Minimax Confidence Intervals for the Sliced Wasserstein Distance
• Borrowing strengh in hierarchical Bayes: Posterior concentration of the Dirichlet base measure
• Posterior contraction of the population polytope in finite admixture models
• Posterior contraction rates for deconvolution of Dirichlet-Laplace mixtures
• Local Minima Structures in Gaussian Mixture Models
• Gaussian Mixture Reduction with Composite Transportation Divergence
• Fast nonparametric near-maximum likelihood estimation of a mixing density
• Bayesian non-parametric inference for $\Lambda$-coalescents: consistency and a parametric method
• Revisiting consistency of a recursive estimator of mixing distributions
• Bayesian mixture models (in)consistency for the number of clusters
• From an Interior Point to a Corner Point: Smart Crossover
• Selective inference after convex clustering with ${\ell }_1$ penalization