An Experimental Study of Prior Dependence in Bayesian Network Structure Learning

Alvaro Henrique Chaim Correia, Cassio P. de Campos, Linda C. van der Gaag
Proceedings of the Eleventh International Symposium on Imprecise Probabilities: Theories and Applications, PMLR 103:78-81, 2019.

Abstract

The Bayesian Dirichlet equivalent uniform (BDeu) function is a popular score to evaluate the goodness of a Bayesian network structure given complete categorical data. Despite its interesting properties, such as likelihood equivalence, it does require a prior expressed via a user-defined parameter known as Equivalent Sample Size (ESS), which significantly affects the final structure. We study conditions to obtain prior independence in BDeu-based structure learning. We show in experiments that the amount of data needed to render the learning robust to different ESS values is prohibitively large, even in big data times.

Cite this Paper

BibTeX
@InProceedings{pmlr-v103-correia19a, title = {An Experimental Study of Prior Dependence in Bayesian Network Structure Learning}, author = {Correia, Alvaro Henrique Chaim and {de Campos}, Cassio P. and {van der Gaag}, Linda C.}, booktitle = {Proceedings of the Eleventh International Symposium on Imprecise Probabilities: Theories and Applications}, pages = {78--81}, year = {2019}, editor = {De Bock, Jasper and de Campos, Cassio P. and de Cooman, Gert and Quaeghebeur, Erik and Wheeler, Gregory}, volume = {103}, series = {Proceedings of Machine Learning Research}, month = {03--06 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v103/correia19a/correia19a.pdf}, url = {https://proceedings.mlr.press/v103/correia19a.html}, abstract = {The Bayesian Dirichlet equivalent uniform (BDeu) function is a popular score to evaluate the goodness of a Bayesian network structure given complete categorical data. Despite its interesting properties, such as likelihood equivalence, it does require a prior expressed via a user-defined parameter known as Equivalent Sample Size (ESS), which significantly affects the final structure. We study conditions to obtain prior independence in BDeu-based structure learning. We show in experiments that the amount of data needed to render the learning robust to different ESS values is prohibitively large, even in big data times.} }
Endnote
%0 Conference Paper %T An Experimental Study of Prior Dependence in Bayesian Network Structure Learning %A Alvaro Henrique Chaim Correia %A Cassio P. de Campos %A Linda C. van der Gaag %B Proceedings of the Eleventh International Symposium on Imprecise Probabilities: Theories and Applications %C Proceedings of Machine Learning Research %D 2019 %E Jasper De Bock %E Cassio P. de Campos %E Gert de Cooman %E Erik Quaeghebeur %E Gregory Wheeler %F pmlr-v103-correia19a %I PMLR %P 78--81 %U https://proceedings.mlr.press/v103/correia19a.html %V 103 %X The Bayesian Dirichlet equivalent uniform (BDeu) function is a popular score to evaluate the goodness of a Bayesian network structure given complete categorical data. Despite its interesting properties, such as likelihood equivalence, it does require a prior expressed via a user-defined parameter known as Equivalent Sample Size (ESS), which significantly affects the final structure. We study conditions to obtain prior independence in BDeu-based structure learning. We show in experiments that the amount of data needed to render the learning robust to different ESS values is prohibitively large, even in big data times.
APA
Correia, A.H.C., de Campos, C.P. & van der Gaag, L.C.. (2019). An Experimental Study of Prior Dependence in Bayesian Network Structure Learning. Proceedings of the Eleventh International Symposium on Imprecise Probabilities: Theories and Applications, in Proceedings of Machine Learning Research 103:78-81 Available from https://proceedings.mlr.press/v103/correia19a.html.

Related Material