Are Gibbs-type priors the most natural generalization of the Dirichlet process?

Abstract

Discrete random probability measures and the exchangeable random partitions they induce are key tools for addressing a variety of estimation and prediction problems in Bayesian inference. Indeed, many popular nonparametric priors, such as the Dirichlet and the Pitman-Yor process priors, select discrete probability distributions almost surely and, therefore, automatically induce exchangeable random partitions. Here we focus on the family of Gibbs-type priors, a recent and elegant generalization of the Dirichlet and the Pitman-Yor process priors. These random probability measures share properties that are appealing both from a theoretical and an applied point of view: (i) they admit an intuitive characterization in terms of their predictive structure justifying their use in terms of a precise assumption on the learning mechanism; (ii) they stand out in terms of mathematical tractability; (iii) they include several interesting special cases besides the Dirichlet and the Pitman-Yor processes. The goal of our paper is to provide a systematic and unified treatment of Gibbs-type priors and highlight their implications for Bayesian nonparametric inference. We will deal with their distributional properties, the resulting estimators, frequentist asymptotic validation and the construction of time-dependent versions. Applications, mainly concerning hierarchical mixture models and species sampling, will serve to convey the main ideas. The intuition inherent to this class of priors and the neat results that can be deduced for it lead one to wonder whether it actually represents the most natural generalization of the Dirichlet process.

arXiv comment5 figures

References in corpus (8)

• Generalized Polya Urn for Time-varying Dirichlet Process Mixtures
• Bayesian semiparametric inference for multivariate doubly-interval-censored data
• Bayesian nonparametric estimators derived from conditional Gibbs structures
• Conditional formulae for Gibbs-type exchangeable random partitions
• Large sample asymptotics for the two-parameter Poisson--Dirichlet process
• Alpha-diversity processes and normalized inverse-Gaussian diffusions
• A Bayesian nonparametric approach to modeling market share dynamics
• Asymptotics for a Bayesian nonparametric estimator of species variety

Cited by in corpus (33)

• Zipf's, Heaps' and Taylor's laws are determined by the expansion into the adjacent possible
• Generalized mixtures of finite mixtures and telescoping sampling
• Infinite Mixtures of Infinite Factor Analysers
• Centered Partition Process: Informative Priors for Clustering
• A marginal sampler for $\sigma$-Stable Poisson-Kingman mixture models
• Looking-backward probabilities for Gibbs-type exchangeable random partitions
• Reducing over-clustering via the powered Chinese restaurant process
• Preferential Attachment and Vertex Arrival Times
• A survey on Bayesian inference for Gaussian mixture model
• Gibbs-type Indian buffet processes
• A nonparametric Bayesian approach to the rare type match problem
• Bayesian nonparametric inference for discovery probabilities: credible intervals and large sample asymptotics
• Nonparametric priors with full-range borrowing of information
• Gibbs sampling for mixtures in order of appearance: the ordered allocation sampler
• Generalized Bayesian Record Linkage and Regression with Exact Error Propagation
• A hybrid sampler for Poisson-Kingman mixture models
• Entropy regularization in probabilistic clustering
• Sampling and Inference for Beta Neutral-to-the-Left Models of Sparse Networks
• Bayesian Nonparametric Models for Multiple Raters: a General Statistical Framework
• Clustering dynamics in a class of normalised generalised gamma dependent priors
• Graph-Aligned Random Partition Model (GARP)
• Bayesian mixture models (in)consistency for the number of clusters
• Learning-augmented count-min sketches via Bayesian nonparametrics
• Consensus Monte Carlo for Random Subsets using Shared Anchors
• A simple proof of Pitman-Yor's Chinese restaurant process from its stick-breaking representation
• Scalable Bayesian Nonparametric Clustering and Classification
• General Bayesian inference schemes in infinite mixture models
• BNPdensity: Bayesian nonparametric mixture modeling in R
• A Dirichlet Process Functional Approach to Heteroscedastic-Consistent Covariance Estimation
• On Johnson's "sufficientness" postulates for features-sampling models
• Bayesian nonparametric modelling of sequential discoveries
• Bayesian analysis of product feature allocation models
• Gibbs Partitions, Riemann-Liouville Fractional Operators, Mittag-Leffler Functions, and Fragmentations Derived From Stable Subordinators