A New Perspective on Learning Context-Specific Independence

Yujia Shen, Arthur Choi, Adnan Darwiche
Proceedings of the 10th International Conference on Probabilistic Graphical Models, PMLR 138:425-436, 2020.

Abstract

Local structure such as context-specific independence (CSI) has received much attention in the probabilistic graphical model (PGM) literature, as it facilitates the modeling of large complex systems, as well as for reasoning with them. In this paper, we provide a new perspective on how to learn CSIs from data. We propose to first learn a functional and parameterized representation of a conditional probability distribution (CPD), such as a neural network. Next, we quantize this continuous function, into an arithmetic circuit representation that facilitates efficient inference. In the first step, we can leverage the many powerful tools that have been developed in the machine learning literature. In the second step, we exploit more recently-developed analytic tools from explainable AI, for the purposes of learning CSIs. Finally, we contrast our approach, empirically and conceptually, with more traditional variable-splitting approaches, that search for CSIs more explicitly.

Cite this Paper


BibTeX
@InProceedings{pmlr-v138-shen20a, title = {A New Perspective on Learning Context-Specific Independence}, author = {Shen, Yujia and Choi, Arthur and Darwiche, Adnan}, booktitle = {Proceedings of the 10th International Conference on Probabilistic Graphical Models}, pages = {425--436}, year = {2020}, editor = {Manfred Jaeger and Thomas Dyhre Nielsen}, volume = {138}, series = {Proceedings of Machine Learning Research}, month = {23--25 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v138/shen20a/shen20a.pdf}, url = { http://proceedings.mlr.press/v138/shen20a.html }, abstract = {Local structure such as context-specific independence (CSI) has received much attention in the probabilistic graphical model (PGM) literature, as it facilitates the modeling of large complex systems, as well as for reasoning with them. In this paper, we provide a new perspective on how to learn CSIs from data. We propose to first learn a functional and parameterized representation of a conditional probability distribution (CPD), such as a neural network. Next, we quantize this continuous function, into an arithmetic circuit representation that facilitates efficient inference. In the first step, we can leverage the many powerful tools that have been developed in the machine learning literature. In the second step, we exploit more recently-developed analytic tools from explainable AI, for the purposes of learning CSIs. Finally, we contrast our approach, empirically and conceptually, with more traditional variable-splitting approaches, that search for CSIs more explicitly.} }
Endnote
%0 Conference Paper %T A New Perspective on Learning Context-Specific Independence %A Yujia Shen %A Arthur Choi %A Adnan Darwiche %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-shen20a %I PMLR %P 425--436 %U http://proceedings.mlr.press/v138/shen20a.html %V 138 %X Local structure such as context-specific independence (CSI) has received much attention in the probabilistic graphical model (PGM) literature, as it facilitates the modeling of large complex systems, as well as for reasoning with them. In this paper, we provide a new perspective on how to learn CSIs from data. We propose to first learn a functional and parameterized representation of a conditional probability distribution (CPD), such as a neural network. Next, we quantize this continuous function, into an arithmetic circuit representation that facilitates efficient inference. In the first step, we can leverage the many powerful tools that have been developed in the machine learning literature. In the second step, we exploit more recently-developed analytic tools from explainable AI, for the purposes of learning CSIs. Finally, we contrast our approach, empirically and conceptually, with more traditional variable-splitting approaches, that search for CSIs more explicitly.
APA
Shen, Y., Choi, A. & Darwiche, A.. (2020). A New Perspective on Learning Context-Specific Independence. Proceedings of the 10th International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 138:425-436 Available from http://proceedings.mlr.press/v138/shen20a.html .

Related Material