Distributed, partially collapsed MCMC for Bayesian Nonparametrics

01/15/2020
by   Avinava Dubey, et al.
33

Bayesian nonparametric (BNP) models provide elegant methods for discovering underlying latent features within a data set, but inference in such models can be slow. We exploit the fact that completely random measures, which commonly used models like the Dirichlet process and the beta-Bernoulli process can be expressed as, are decomposable into independent sub-measures. We use this decomposition to partition the latent measure into a finite measure containing only instantiated components, and an infinite measure containing all other components. We then select different inference algorithms for the two components: uncollapsed samplers mix well on the finite measure, while collapsed samplers mix well on the infinite, sparsely occupied tail. The resulting hybrid algorithm can be applied to a wide class of models, and can be easily distributed to allow scalable inference without sacrificing asymptotic convergence guarantees.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/09/2017

Parallel Markov Chain Monte Carlo for the Indian Buffet Process

Indian Buffet Process based models are an elegant way for discovering un...
research
09/22/2020

Independent finite approximations for Bayesian nonparametric inference: construction, error bounds, and practical implications

Bayesian nonparametrics based on completely random measures (CRMs) offer...
research
05/26/2019

A unified construction for series representations and finite approximations of completely random measures

Infinite-activity completely random measures (CRMs) have become importan...
research
05/17/2020

Truncated Self-Product Measures in Edge-Exchangeable Networks

Edge-exchangeable probabilistic network models generate edges as an i.i....
research
06/24/2020

Slice Sampling for General Completely Random Measures

Completely random measures provide a principled approach to creating fle...
research
08/17/2019

Black-box constructions for exchangeable sequences of random multisets

We develop constructions for exchangeable sequences of point processes t...
research
05/09/2012

The Infinite Latent Events Model

We present the Infinite Latent Events Model, a nonparametric hierarchica...

Please sign up or login with your details

Forgot password? Click here to reset