Cascading Sum-Product Networks using Robustness

Diarmaid Conaty, Jesús Martínez Del Rincon, Cassio P. De Campos
Proceedings of the Ninth International Conference on Probabilistic Graphical Models, PMLR 72:73-84, 2018.

Abstract

Sum-product networks are an increasingly popular family of probabilistic graphical models for which marginal inference can be performed in polynomial time. They have been shown to achieve state-of-the-art performance in several tasks. When learning sum-product networks from scarce data, the obtained model may be prone to robustness issues. In particular, small variations of parameters could lead to different conclusions. We discuss the characteristics of sum-product networks as classifiers and study the robustness of them with respect to their parameters. Using a robustness measure to identify (possibly) unreliable decisions, we build a hierarchical approach where the classification task is deferred to another model if the outcome is deemed unreliable. We apply this approach on benchmark classification tasks and experiments show that the robustness measure can be a meaningful manner to improve classification accuracy.

Cite this Paper


BibTeX
@InProceedings{pmlr-v72-conaty18a, title = {Cascading Sum-Product Networks using Robustness}, author = {Conaty, Diarmaid and Mart\'{i}nez Del Rincon, Jes\'{u}s and De Campos, Cassio P.}, booktitle = {Proceedings of the Ninth International Conference on Probabilistic Graphical Models}, pages = {73--84}, year = {2018}, editor = {Kratochvíl, Václav and Studený, Milan}, volume = {72}, series = {Proceedings of Machine Learning Research}, month = {11--14 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v72/conaty18a/conaty18a.pdf}, url = {https://proceedings.mlr.press/v72/conaty18a.html}, abstract = {Sum-product networks are an increasingly popular family of probabilistic graphical models for which marginal inference can be performed in polynomial time. They have been shown to achieve state-of-the-art performance in several tasks. When learning sum-product networks from scarce data, the obtained model may be prone to robustness issues. In particular, small variations of parameters could lead to different conclusions. We discuss the characteristics of sum-product networks as classifiers and study the robustness of them with respect to their parameters. Using a robustness measure to identify (possibly) unreliable decisions, we build a hierarchical approach where the classification task is deferred to another model if the outcome is deemed unreliable. We apply this approach on benchmark classification tasks and experiments show that the robustness measure can be a meaningful manner to improve classification accuracy.} }
Endnote
%0 Conference Paper %T Cascading Sum-Product Networks using Robustness %A Diarmaid Conaty %A Jesús Martínez Del Rincon %A Cassio P. De Campos %B Proceedings of the Ninth International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2018 %E Václav Kratochvíl %E Milan Studený %F pmlr-v72-conaty18a %I PMLR %P 73--84 %U https://proceedings.mlr.press/v72/conaty18a.html %V 72 %X Sum-product networks are an increasingly popular family of probabilistic graphical models for which marginal inference can be performed in polynomial time. They have been shown to achieve state-of-the-art performance in several tasks. When learning sum-product networks from scarce data, the obtained model may be prone to robustness issues. In particular, small variations of parameters could lead to different conclusions. We discuss the characteristics of sum-product networks as classifiers and study the robustness of them with respect to their parameters. Using a robustness measure to identify (possibly) unreliable decisions, we build a hierarchical approach where the classification task is deferred to another model if the outcome is deemed unreliable. We apply this approach on benchmark classification tasks and experiments show that the robustness measure can be a meaningful manner to improve classification accuracy.
APA
Conaty, D., Martínez Del Rincon, J. & De Campos, C.P.. (2018). Cascading Sum-Product Networks using Robustness. Proceedings of the Ninth International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 72:73-84 Available from https://proceedings.mlr.press/v72/conaty18a.html.

Related Material