The Efficient Propagation of Arbitrary Subsets of Beliefs in Discrete-Valued Bayesian Networks

Duncan Smith
Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics, PMLR R3:272-277, 2001.

Abstract

The paper describes an approach for propagating arbitrary subsets of beliefs in Bayesian Belief Networks. The method is based on a multiple message passing scheme in junction trees. A hybrid tree structure is introduced, both for the propagation of evidence and as an efficiently permutable representation of a decomposable graph. The use of maximal prime subgraph decompositions and tree permutations to reduce computational cost is demonstrated.

Cite this Paper


BibTeX
@InProceedings{pmlr-vR3-smith01a, title = {The Efficient Propagation of Arbitrary Subsets of Beliefs in Discrete-Valued Bayesian Networks}, author = {Smith, Duncan}, booktitle = {Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics}, pages = {272--277}, year = {2001}, editor = {Richardson, Thomas S. and Jaakkola, Tommi S.}, volume = {R3}, series = {Proceedings of Machine Learning Research}, month = {04--07 Jan}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/r3/smith01a/smith01a.pdf}, url = {https://proceedings.mlr.press/r3/smith01a.html}, abstract = {The paper describes an approach for propagating arbitrary subsets of beliefs in Bayesian Belief Networks. The method is based on a multiple message passing scheme in junction trees. A hybrid tree structure is introduced, both for the propagation of evidence and as an efficiently permutable representation of a decomposable graph. The use of maximal prime subgraph decompositions and tree permutations to reduce computational cost is demonstrated.}, note = {Reissued by PMLR on 31 March 2021.} }
Endnote
%0 Conference Paper %T The Efficient Propagation of Arbitrary Subsets of Beliefs in Discrete-Valued Bayesian Networks %A Duncan Smith %B Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2001 %E Thomas S. Richardson %E Tommi S. Jaakkola %F pmlr-vR3-smith01a %I PMLR %P 272--277 %U https://proceedings.mlr.press/r3/smith01a.html %V R3 %X The paper describes an approach for propagating arbitrary subsets of beliefs in Bayesian Belief Networks. The method is based on a multiple message passing scheme in junction trees. A hybrid tree structure is introduced, both for the propagation of evidence and as an efficiently permutable representation of a decomposable graph. The use of maximal prime subgraph decompositions and tree permutations to reduce computational cost is demonstrated. %Z Reissued by PMLR on 31 March 2021.
APA
Smith, D.. (2001). The Efficient Propagation of Arbitrary Subsets of Beliefs in Discrete-Valued Bayesian Networks. Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research R3:272-277 Available from https://proceedings.mlr.press/r3/smith01a.html. Reissued by PMLR on 31 March 2021.

Related Material