Approximating bounded tree-width Bayesian network classifiers with OBDD

Karine Chubarian, György Turán
Proceedings of the 10th International Conference on Probabilistic Graphical Models, PMLR 138:113-124, 2020.

Abstract

It is shown that Bayesian network classifiers of tree-width $k$ have an OBDD approximation computable in polynomial time in the parameters, for every fixed $k$. This is shown by approximating a polynomial threshold function representing the classifier. The approximation error can be measured with respect to any distribution which can be approximated by a mixture of bounded width distributions. This includes the input distribution of the classifier.

Cite this Paper

BibTeX
@InProceedings{pmlr-v138-chubarian20a, title = { Approximating bounded tree-width Bayesian network classifiers with OBDD}, author = {Chubarian, Karine and Tur\'an, Gy\"orgy}, booktitle = {Proceedings of the 10th International Conference on Probabilistic Graphical Models}, pages = {113--124}, year = {2020}, editor = {Jaeger, Manfred and Nielsen, Thomas Dyhre}, volume = {138}, series = {Proceedings of Machine Learning Research}, month = {23--25 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v138/chubarian20a/chubarian20a.pdf}, url = {https://proceedings.mlr.press/v138/chubarian20a.html}, abstract = {It is shown that Bayesian network classifiers of tree-width $k$ have an OBDD approximation computable in polynomial time in the parameters, for every fixed $k$. This is shown by approximating a polynomial threshold function representing the classifier. The approximation error can be measured with respect to any distribution which can be approximated by a mixture of bounded width distributions. This includes the input distribution of the classifier.} }
Endnote
%0 Conference Paper %T Approximating bounded tree-width Bayesian network classifiers with OBDD %A Karine Chubarian %A György Turán %B Proceedings of the 10th International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2020 %E Manfred Jaeger %E Thomas Dyhre Nielsen %F pmlr-v138-chubarian20a %I PMLR %P 113--124 %U https://proceedings.mlr.press/v138/chubarian20a.html %V 138 %X It is shown that Bayesian network classifiers of tree-width $k$ have an OBDD approximation computable in polynomial time in the parameters, for every fixed $k$. This is shown by approximating a polynomial threshold function representing the classifier. The approximation error can be measured with respect to any distribution which can be approximated by a mixture of bounded width distributions. This includes the input distribution of the classifier.
APA
Chubarian, K. & Turán, G.. (2020). Approximating bounded tree-width Bayesian network classifiers with OBDD. Proceedings of the 10th International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 138:113-124 Available from https://proceedings.mlr.press/v138/chubarian20a.html.

Related Material