English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Evaluation of marginal likelihoods via the density of states

Habeck, M. (2012). Evaluation of marginal likelihoods via the density of states. Journal of Machine Learning Research: Workshop and Conference Proceedings, 22, 486-494.

Item is

Files

show Files

Locators

show

Creators

show
hide
 Creators:
Habeck, M1, Author           
Lawrence, N, Editor
Girolami, M, Editor
Affiliations:
1Dept. Empirical Inference, Max Planck Institute for Intelligent Systems, Max Planck Society, ou_1497647              

Content

show
hide
Free keywords: Abt. Schölkopf
 Abstract: Bayesian model comparison involves the evaluation of the marginal likelihood, the expectation of the likelihood under the prior distribution. Typically, this high-dimensional integral over all model parameters is approximated using Markov chain Monte Carlo methods. Thermodynamic integration is a popular method to estimate the marginal likelihood by using samples from annealed posteriors. Here we show that there exists a robust and flexible alternative. The new method estimates the density of states, which counts the number of states associated with a particular value of the likelihood. If the density of states is known, computation of the marginal likelihood reduces to a one- dimensional integral. We outline a maximum likelihood procedure to estimate the density of states from annealed posterior samples. We apply our method to various likelihoods and show that it is superior to thermodynamic integration in that it is more flexible with regard to the annealing schedule and the family of bridging distributions. Finally, we discuss the relation of our method with Skilling's nested sampling.

Details

show
hide
Language(s):
 Dates: 2012
 Publication Status: Issued
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: BibTex Citekey: Habeck2012_3
 Degree: -

Event

show
hide
Title: AISTATS 2012
Place of Event: -
Start-/End Date: -

Legal Case

show

Project information

show

Source 1

show
hide
Title: Journal of Machine Learning Research: Workshop and Conference Proceedings
  Abbreviation : JMLR: W&P
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: JMLR: WCP 22
Pages: - Volume / Issue: 22 Sequence Number: - Start / End Page: 486 - 494 Identifier: -