Dissertation/Thesis Abstract

General Purpose MCMC Sampling for Bayesian Model Averaging
by Boyles, Levi Beinarauskas, Ph.D., University of California, Irvine, 2014, 129; 3631086
Abstract (Summary)

In this thesis we explore the problem of inference for Bayesian model averaging. Many popular topics in Bayesian analysis, such as Bayesian nonparametrics, can be cast as model averaging problems. Model averaging problems offer unique difficulties for inference, as the parameter space is not fixed, and may be infinite. As such, there is little existing work on general purpose MCMC algorithms in this area. We introduce a new MCMC sampler, which we call Retrospective Jump sampling, that is suitable for general purpose model averaging. In the development of Retrospective Jump, some practical issues arise in the need for a MCMC sampler for finite dimensions that is suitable for multimodal target densities; we introduce Refractive Sampling as a sampler suitable in this regard. Finally, we evaluate Retrospective Jump on several model averaging and Bayesian nonparametric problems, and develop a novel latent feature model with hierarchical column structure which uses Retrospective Jump for inference.

Indexing (document details)
Advisor: Welling, Max
Commitee: Shahbaba, Babak, Smyth, Padhraic
School: University of California, Irvine
Department: Computer Science
School Location: United States -- California
Source: DAI-B 75/11(E), Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Computer Engineering, Computer science
Keywords: Averaging, Bayesian model, MCMC, Retrospective jump sampling
Publication Number: 3631086
ISBN: 9781321093704
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest