Markov chain Monte Carlo using tree-based priors on model structure.

Angelopoulos, Nicos ORCID: https://orcid.org/0000-0002-7507-9177 and Cussens, James 2001. Markov chain Monte Carlo using tree-based priors on model structure. Presented at: 17th The Conference on Uncertainty in Artificial Intelligence (UAI 2001), Seattle, WA, USA, 2-5 August 2001. Published in: Breese, Jack and Koller, Daphne eds. UAI'01: Proceedings of the Seventeenth conference on Uncertainty in artificial intelligence. San Francisco, USA: Morgan Kaufmann Publishers Inc., 10.5555/2074022.2074025

Full text not available from this repository.

Abstract

We present a general framework for defining priors on model structure and sampling from the posterior using the Metropolis-Hastings algorithm. The key ideas are that structure priors are defined via a probability tree and that the proposal distribution for the Metropolis-Hastings algorithm is defined using the prior, thereby defining a cheaply computable acceptance probability. We have applied this approach to Bayesian net structure learning using a number of priors and proposal distributions. Our results show that these must be chosen appropriately for this approach to be successful.

Item Type:	Conference or Workshop Item (Paper)
Status:	Published
Schools:	Medicine
Publisher:	Morgan Kaufmann Publishers Inc.
ISBN:	9781558608009
Date of Acceptance:	31 May 2001
Last Modified:	04 Jan 2023 02:20
URI:	https://orca.cardiff.ac.uk/id/eprint/134004

Actions (repository staff only)

Edit Item