Tempering for Bayesian C&RT

J. Cussens; S. Wrobel; N. Angelopoulos; L. De Raedt

doi:10.1145/1102351.1102354

Tempering for Bayesian C&RT

J. Cussens, S. Wrobel (Editor), N. Angelopoulos, L. De Raedt (Editor)

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper concerns the experimental assessment of tempering as a technique for improving Bayesian inference for C&RT models. Full Bayesian inference requires the computation of a posterior over all possible trees. Since exact computation is not possible Markov chain Monte Carlo (MCMC) methods are used to produce an approximation. C&RT posteriors have many local modes: tempering aims to prevent the Markov chain getting stuck in these modes. Our results show that a clear improvement is achieved using tempering.

Original language	English
Title of host publication	Proceedings of the 22nd International Conference on Machine Learning
Pages	17-24
Number of pages	7
DOIs	https://doi.org/10.1145/1102351.1102354
Publication status	Published - 2005
Event	ICML 2005 - Bonn, Germany Duration: 7 Aug 2005 → 11 Aug 2005

Publication series

Name	ACM International Conference Proceeding Series
Volume	19

Conference

Conference	ICML 2005
City	Bonn, Germany
Period	7/08/05 → 11/08/05

Access to Document

10.1145/1102351.1102354

http://dx.doi.org/10.1145/1102351.1102354

Cite this

@inproceedings{67a44b89edf4408a998e3f646ab36024,

title = "Tempering for Bayesian C&RT",

abstract = "This paper concerns the experimental assessment of tempering as a technique for improving Bayesian inference for C&RT models. Full Bayesian inference requires the computation of a posterior over all possible trees. Since exact computation is not possible Markov chain Monte Carlo (MCMC) methods are used to produce an approximation. C&RT posteriors have many local modes: tempering aims to prevent the Markov chain getting stuck in these modes. Our results show that a clear improvement is achieved using tempering.",

author = "J. Cussens and S. Wrobel and N. Angelopoulos and {De Raedt}, L.",

year = "2005",

doi = "10.1145/1102351.1102354",

language = "English",

isbn = "1-59593-180-5",

series = "ACM International Conference Proceeding Series",

pages = "17--24",

booktitle = "Proceedings of the 22nd International Conference on Machine Learning",

note = "ICML 2005 ; Conference date: 07-08-2005 Through 11-08-2005",

}

TY - GEN

T1 - Tempering for Bayesian C&RT

AU - Cussens, J.

AU - Angelopoulos, N.

A2 - Wrobel, S.

A2 - De Raedt, L.

PY - 2005

Y1 - 2005

N2 - This paper concerns the experimental assessment of tempering as a technique for improving Bayesian inference for C&RT models. Full Bayesian inference requires the computation of a posterior over all possible trees. Since exact computation is not possible Markov chain Monte Carlo (MCMC) methods are used to produce an approximation. C&RT posteriors have many local modes: tempering aims to prevent the Markov chain getting stuck in these modes. Our results show that a clear improvement is achieved using tempering.

AB - This paper concerns the experimental assessment of tempering as a technique for improving Bayesian inference for C&RT models. Full Bayesian inference requires the computation of a posterior over all possible trees. Since exact computation is not possible Markov chain Monte Carlo (MCMC) methods are used to produce an approximation. C&RT posteriors have many local modes: tempering aims to prevent the Markov chain getting stuck in these modes. Our results show that a clear improvement is achieved using tempering.

U2 - 10.1145/1102351.1102354

DO - 10.1145/1102351.1102354

M3 - Conference contribution

SN - 1-59593-180-5

T3 - ACM International Conference Proceeding Series

SP - 17

EP - 24

BT - Proceedings of the 22nd International Conference on Machine Learning

T2 - ICML 2005

Y2 - 7 August 2005 through 11 August 2005

ER -