Learning Tractable Multidimensional Bayesian Network Classifiers

Marco Benjumeda, Concha Bielza, Pedro Larrañaga
Proceedings of the Eighth International Conference on Probabilistic Graphical Models, PMLR 52:13-24, 2016.

Abstract

Multidimensional classification has become one of the most relevant topics in view of the many domains that require a vector of class values to be assigned to a vector of given features. The popularity of multidimensional Bayesian network classifiers has increased in the last few years due to their expressive power and the existence of methods for learning different families of these models. The problem with this approach is that the computational cost of using the learned models is usually high, especially if there are a lot of class variables. Class-bridge decomposability means that the multidimensional classification problem can be divided into multiple subproblems for these models. In this paper, we prove that class-bridge decomposability can also be used to guarantee the tractability of the models. We also propose a strategy for efficiently bounding their inference complexity, providing a simple learning method with an order-based search that obtains tractable multidimensional Bayesian network classifiers. Experimental results show that our approach is competitive with other methods in the state of the art and ensures the tractability of the learned models.

Cite this Paper

BibTeX
@InProceedings{pmlr-v52-benjumeda16, title = {Learning Tractable Multidimensional {B}ayesian Network Classifiers}, author = {Benjumeda, Marco and Bielza, Concha and Larrañaga, Pedro}, booktitle = {Proceedings of the Eighth International Conference on Probabilistic Graphical Models}, pages = {13--24}, 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/benjumeda16.pdf}, url = {https://proceedings.mlr.press/v52/benjumeda16.html}, abstract = {Multidimensional classification has become one of the most relevant topics in view of the many domains that require a vector of class values to be assigned to a vector of given features. The popularity of multidimensional Bayesian network classifiers has increased in the last few years due to their expressive power and the existence of methods for learning different families of these models. The problem with this approach is that the computational cost of using the learned models is usually high, especially if there are a lot of class variables. Class-bridge decomposability means that the multidimensional classification problem can be divided into multiple subproblems for these models. In this paper, we prove that class-bridge decomposability can also be used to guarantee the tractability of the models. We also propose a strategy for efficiently bounding their inference complexity, providing a simple learning method with an order-based search that obtains tractable multidimensional Bayesian network classifiers. Experimental results show that our approach is competitive with other methods in the state of the art and ensures the tractability of the learned models.} }
Endnote
%0 Conference Paper %T Learning Tractable Multidimensional Bayesian Network Classifiers %A Marco Benjumeda %A Concha Bielza %A Pedro Larrañaga %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-benjumeda16 %I PMLR %P 13--24 %U https://proceedings.mlr.press/v52/benjumeda16.html %V 52 %X Multidimensional classification has become one of the most relevant topics in view of the many domains that require a vector of class values to be assigned to a vector of given features. The popularity of multidimensional Bayesian network classifiers has increased in the last few years due to their expressive power and the existence of methods for learning different families of these models. The problem with this approach is that the computational cost of using the learned models is usually high, especially if there are a lot of class variables. Class-bridge decomposability means that the multidimensional classification problem can be divided into multiple subproblems for these models. In this paper, we prove that class-bridge decomposability can also be used to guarantee the tractability of the models. We also propose a strategy for efficiently bounding their inference complexity, providing a simple learning method with an order-based search that obtains tractable multidimensional Bayesian network classifiers. Experimental results show that our approach is competitive with other methods in the state of the art and ensures the tractability of the learned models.
RIS
TY - CPAPER TI - Learning Tractable Multidimensional Bayesian Network Classifiers AU - Marco Benjumeda AU - Concha Bielza AU - Pedro Larrañaga 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-benjumeda16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 52 SP - 13 EP - 24 L1 - http://proceedings.mlr.press/v52/benjumeda16.pdf UR - https://proceedings.mlr.press/v52/benjumeda16.html AB - Multidimensional classification has become one of the most relevant topics in view of the many domains that require a vector of class values to be assigned to a vector of given features. The popularity of multidimensional Bayesian network classifiers has increased in the last few years due to their expressive power and the existence of methods for learning different families of these models. The problem with this approach is that the computational cost of using the learned models is usually high, especially if there are a lot of class variables. Class-bridge decomposability means that the multidimensional classification problem can be divided into multiple subproblems for these models. In this paper, we prove that class-bridge decomposability can also be used to guarantee the tractability of the models. We also propose a strategy for efficiently bounding their inference complexity, providing a simple learning method with an order-based search that obtains tractable multidimensional Bayesian network classifiers. Experimental results show that our approach is competitive with other methods in the state of the art and ensures the tractability of the learned models. ER -
APA
Benjumeda, M., Bielza, C. & Larrañaga, P.. (2016). Learning Tractable Multidimensional Bayesian Network Classifiers. Proceedings of the Eighth International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 52:13-24 Available from https://proceedings.mlr.press/v52/benjumeda16.html.

Related Material