On Bayesian Network Inference with Simple Propagation

Cory J. Butz, Jhonatan S. Oliveira, André E. dos Santos, Anders L. Madsen
; Proceedings of the Eighth International Conference on Probabilistic Graphical Models, PMLR 52:62-73, 2016.

Abstract

\emphSimple Propagation (SP) was recently proposed as a new join tree propagation algorithm for exact inference in discrete Bayesian networks and empirically shown to be faster than \emphLazy Propagation (LP) when applied on optimal (or close to) join trees built from real-world and benchmark Bayesian networks. This paper extends SP in two directions. First, we propose and empirically evaluate eight heuristics for determining elimination orderings in SP. Second, we show that the relevant potentials in SP are precisely those in LP.

Cite this Paper


BibTeX
@InProceedings{pmlr-v52-butz16a, title = {On {B}ayesian Network Inference with Simple Propagation}, author = {Cory J. Butz and Jhonatan S. Oliveira and André E. dos Santos and Anders L. Madsen}, booktitle = {Proceedings of the Eighth International Conference on Probabilistic Graphical Models}, pages = {62--73}, year = {2016}, editor = {Alessandro Antonucci and Giorgio Corani and Cassio Polpo Campos}}, volume = {52}, series = {Proceedings of Machine Learning Research}, address = {Lugano, Switzerland}, month = {06--09 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v52/butz16a.pdf}, url = {http://proceedings.mlr.press/v52/butz16a.html}, abstract = {\emphSimple Propagation (SP) was recently proposed as a new join tree propagation algorithm for exact inference in discrete Bayesian networks and empirically shown to be faster than \emphLazy Propagation (LP) when applied on optimal (or close to) join trees built from real-world and benchmark Bayesian networks. This paper extends SP in two directions. First, we propose and empirically evaluate eight heuristics for determining elimination orderings in SP. Second, we show that the relevant potentials in SP are precisely those in LP.} }
Endnote
%0 Conference Paper %T On Bayesian Network Inference with Simple Propagation %A Cory J. Butz %A Jhonatan S. Oliveira %A André E. dos Santos %A Anders L. Madsen %B Proceedings of the Eighth International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2016 %E Alessandro Antonucci %E Giorgio Corani %E Cassio Polpo Campos} %F pmlr-v52-butz16a %I PMLR %J Proceedings of Machine Learning Research %P 62--73 %U http://proceedings.mlr.press %V 52 %W PMLR %X \emphSimple Propagation (SP) was recently proposed as a new join tree propagation algorithm for exact inference in discrete Bayesian networks and empirically shown to be faster than \emphLazy Propagation (LP) when applied on optimal (or close to) join trees built from real-world and benchmark Bayesian networks. This paper extends SP in two directions. First, we propose and empirically evaluate eight heuristics for determining elimination orderings in SP. Second, we show that the relevant potentials in SP are precisely those in LP.
RIS
TY - CPAPER TI - On Bayesian Network Inference with Simple Propagation AU - Cory J. Butz AU - Jhonatan S. Oliveira AU - André E. dos Santos AU - Anders L. Madsen BT - Proceedings of the Eighth International Conference on Probabilistic Graphical Models PY - 2016/08/15 DA - 2016/08/15 ED - Alessandro Antonucci ED - Giorgio Corani ED - Cassio Polpo Campos} ID - pmlr-v52-butz16a PB - PMLR SP - 62 DP - PMLR EP - 73 L1 - http://proceedings.mlr.press/v52/butz16a.pdf UR - http://proceedings.mlr.press/v52/butz16a.html AB - \emphSimple Propagation (SP) was recently proposed as a new join tree propagation algorithm for exact inference in discrete Bayesian networks and empirically shown to be faster than \emphLazy Propagation (LP) when applied on optimal (or close to) join trees built from real-world and benchmark Bayesian networks. This paper extends SP in two directions. First, we propose and empirically evaluate eight heuristics for determining elimination orderings in SP. Second, we show that the relevant potentials in SP are precisely those in LP. ER -
APA
Butz, C.J., Oliveira, J.S., Santos, A.E.d. & Madsen, A.L.. (2016). On Bayesian Network Inference with Simple Propagation. Proceedings of the Eighth International Conference on Probabilistic Graphical Models, in PMLR 52:62-73

Related Material