Relevant Path Separation: A Faster Method for Testing Independencies in Bayesian Networks

Cory J. Butz, André E. dos Santos, Jhonatan S. Oliveira
Proceedings of the Eighth International Conference on Probabilistic Graphical Models, PMLR 52:74-85, 2016.

Abstract

\emphDirected separation (d-separation) played a fundamental role in the founding of \emphBayesian networks (BNs) and continues to be useful today in a wide range of applications. Given an independence to be tested, current implementations of d-separation explore the \emphactive part of a BN. On the other hand, an overlooked property of d-separation implies that d-separation need only consider the \emphrelevant part of a BN. We propose a new method for testing independencies in BNs, called \emphrelevant path separation (rp-separation), which explores the intersection between the active and relevant parts of a BN. Favourable experimental results are reported.

Cite this Paper

BibTeX
@InProceedings{pmlr-v52-butz16b, title = {Relevant Path Separation: A Faster Method for Testing Independencies in {B}ayesian Networks}, author = {Butz, Cory J. and Santos, André E. dos and Oliveira, Jhonatan S.}, booktitle = {Proceedings of the Eighth International Conference on Probabilistic Graphical Models}, pages = {74--85}, year = {2016}, editor = {Antonucci, Alessandro and Corani, Giorgio and Campos}, Cassio Polpo}, volume = {52}, series = {Proceedings of Machine Learning Research}, address = {Lugano, Switzerland}, month = {06--09 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v52/butz16b.pdf}, url = {https://proceedings.mlr.press/v52/butz16b.html}, abstract = {\emphDirected separation (d-separation) played a fundamental role in the founding of \emphBayesian networks (BNs) and continues to be useful today in a wide range of applications. Given an independence to be tested, current implementations of d-separation explore the \emphactive part of a BN. On the other hand, an overlooked property of d-separation implies that d-separation need only consider the \emphrelevant part of a BN. We propose a new method for testing independencies in BNs, called \emphrelevant path separation (rp-separation), which explores the intersection between the active and relevant parts of a BN. Favourable experimental results are reported.} }
Endnote
%0 Conference Paper %T Relevant Path Separation: A Faster Method for Testing Independencies in Bayesian Networks %A Cory J. Butz %A André E. dos Santos %A Jhonatan S. Oliveira %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-butz16b %I PMLR %P 74--85 %U https://proceedings.mlr.press/v52/butz16b.html %V 52 %X \emphDirected separation (d-separation) played a fundamental role in the founding of \emphBayesian networks (BNs) and continues to be useful today in a wide range of applications. Given an independence to be tested, current implementations of d-separation explore the \emphactive part of a BN. On the other hand, an overlooked property of d-separation implies that d-separation need only consider the \emphrelevant part of a BN. We propose a new method for testing independencies in BNs, called \emphrelevant path separation (rp-separation), which explores the intersection between the active and relevant parts of a BN. Favourable experimental results are reported.
RIS
TY - CPAPER TI - Relevant Path Separation: A Faster Method for Testing Independencies in Bayesian Networks AU - Cory J. Butz AU - André E. dos Santos AU - Jhonatan S. Oliveira BT - Proceedings of the Eighth International Conference on Probabilistic Graphical Models DA - 2016/08/15 ED - Alessandro Antonucci ED - Giorgio Corani ED - Cassio Polpo Campos} ID - pmlr-v52-butz16b PB - PMLR DP - Proceedings of Machine Learning Research VL - 52 SP - 74 EP - 85 L1 - http://proceedings.mlr.press/v52/butz16b.pdf UR - https://proceedings.mlr.press/v52/butz16b.html AB - \emphDirected separation (d-separation) played a fundamental role in the founding of \emphBayesian networks (BNs) and continues to be useful today in a wide range of applications. Given an independence to be tested, current implementations of d-separation explore the \emphactive part of a BN. On the other hand, an overlooked property of d-separation implies that d-separation need only consider the \emphrelevant part of a BN. We propose a new method for testing independencies in BNs, called \emphrelevant path separation (rp-separation), which explores the intersection between the active and relevant parts of a BN. Favourable experimental results are reported. ER -
APA
Butz, C.J., Santos, A.E.d. & Oliveira, J.S.. (2016). Relevant Path Separation: A Faster Method for Testing Independencies in Bayesian Networks. Proceedings of the Eighth International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 52:74-85 Available from https://proceedings.mlr.press/v52/butz16b.html.

Related Material