Timezone: »
We present a novel MCMC sampler for Dirichlet process mixture models that can be used for conjugate or non-conjugate prior distributions. The proposed sampler can be massively parallelized to achieve significant computational gains. A non-ergodic restricted Gibbs iteration is mixed with split/merge proposals to produce a valid sampler. Each regular cluster is augmented with two sub-clusters to construct likely split moves. Unlike many previous parallel samplers, the proposed sampler accurately enforces the correct stationary distribution of the Markov chain without the need for approximate models. Empirical results illustrate that the new sampler exhibits better convergence properties than current methods.
Author Information
Jason Chang (Google)
John Fisher III (MIT)
More from the Same Authors
-
2022 : Posterior Consistency for Gaussian Process Surrogate Models with Generalized Observations »
Rujian Chen · John Fisher III -
2020 Poster: Sequential Bayesian Experimental Design with Variable Cost Structure »
Sue Zheng · David Hayden · Jason Pacheco · John Fisher III -
2020 Poster: Belief-Dependent Macro-Action Discovery in POMDPs using the Value of Information »
Genevieve Flaspohler · Nick Roy · John Fisher III -
2015 Poster: Streaming, Distributed Variational Inference for Bayesian Nonparametrics »
Trevor Campbell · Julian Straub · John Fisher III · Jonathan How -
2015 Poster: Probabilistic Variational Bounds for Graphical Models »
Qiang Liu · John Fisher III · Alexander Ihler -
2014 Poster: Coresets for k-Segmentation of Streaming Data »
Guy Rosman · Mikhail Volkov · Dan Feldman · John Fisher III · Daniela Rus -
2014 Poster: Parallel Sampling of HDPs using Sub-Cluster Splits »
Jason Chang · John Fisher III -
2012 Workshop: Bayesian Nonparametric Models For Reliable Planning And Decision-Making Under Uncertainty »
Jonathan How · Lawrence Carin · John Fisher III · Michael Jordan · Alborz Geramifard -
2012 Poster: Coupling Nonparametric Mixtures via Latent Dirichlet Processes »
Dahua Lin · John Fisher III -
2010 Oral: Construction of Dependent Dirichlet Processes based on Poisson Processes »
Dahua Lin · Eric Grimson · John Fisher III -
2010 Poster: Construction of Dependent Dirichlet Processes based on Poisson Processes »
Dahua Lin · Eric Grimson · John Fisher III