Propagation using Chain Event Graphs

Peter A. Thwaites, Jim Q. Smith, Robert G. Cowell
Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence, PMLR R6:546-553, 2008.

Abstract

A Chain Event Graph (CEG) is a graphical model which is designed to embody conditional independencies in problems whose state spaces are highly asymmetric and do not admit a natural product structure. In this paper we present a probability propagation algorithm which uses the topology of the CEG to build a transporter CEG. Intriguingly, the transporter CEG is directly analogous to the triangulated Bayesian Network (BN) in the more conventional junction tree propagation algorithms used with BNs. The propagation method uses factorization formulae also analogous to (but different from) the ones using potentials on cliques and separators of the BN. It appears that the methods will be typically more efficient than the BN algorithms when applied to contexts where there is significant asymmetry present.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR6-thwaites08a, title = {Propagation using Chain Event Graphs}, author = {Thwaites, Peter A. and Smith, Jim Q. and Cowell, Robert G.}, booktitle = {Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence}, pages = {546--553}, year = {2008}, editor = {McAllester, David A. and Myllymäki, Petri}, volume = {R6}, series = {Proceedings of Machine Learning Research}, month = {09--12 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r6/main/assets/thwaites08a/thwaites08a.pdf}, url = {https://proceedings.mlr.press/r6/thwaites08a.html}, abstract = {A Chain Event Graph (CEG) is a graphical model which is designed to embody conditional independencies in problems whose state spaces are highly asymmetric and do not admit a natural product structure. In this paper we present a probability propagation algorithm which uses the topology of the CEG to build a transporter CEG. Intriguingly, the transporter CEG is directly analogous to the triangulated Bayesian Network (BN) in the more conventional junction tree propagation algorithms used with BNs. The propagation method uses factorization formulae also analogous to (but different from) the ones using potentials on cliques and separators of the BN. It appears that the methods will be typically more efficient than the BN algorithms when applied to contexts where there is significant asymmetry present.}, note = {Reissued by PMLR on 09 October 2024.} }
Endnote
%0 Conference Paper %T Propagation using Chain Event Graphs %A Peter A. Thwaites %A Jim Q. Smith %A Robert G. Cowell %B Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2008 %E David A. McAllester %E Petri Myllymäki %F pmlr-vR6-thwaites08a %I PMLR %P 546--553 %U https://proceedings.mlr.press/r6/thwaites08a.html %V R6 %X A Chain Event Graph (CEG) is a graphical model which is designed to embody conditional independencies in problems whose state spaces are highly asymmetric and do not admit a natural product structure. In this paper we present a probability propagation algorithm which uses the topology of the CEG to build a transporter CEG. Intriguingly, the transporter CEG is directly analogous to the triangulated Bayesian Network (BN) in the more conventional junction tree propagation algorithms used with BNs. The propagation method uses factorization formulae also analogous to (but different from) the ones using potentials on cliques and separators of the BN. It appears that the methods will be typically more efficient than the BN algorithms when applied to contexts where there is significant asymmetry present. %Z Reissued by PMLR on 09 October 2024.
APA
Thwaites, P.A., Smith, J.Q. & Cowell, R.G.. (2008). Propagation using Chain Event Graphs. Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R6:546-553 Available from https://proceedings.mlr.press/r6/thwaites08a.html. Reissued by PMLR on 09 October 2024.

Related Material