Inferring Block Structure of Graphical Models in Exponential Families

Siqi Sun, Hai Wang, Jinbo Xu
Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, PMLR 38:939-947, 2015.

Abstract

Learning the structure of a graphical model is a fundamental problem and it is used extensively to infer the relationship between random variables. In many real world applications, we usually have some prior knowledge about the underlying graph structure, such as degree distribution and block structure. In this paper, we propose a novel generative model for describing the block structure in general exponential families, and optimize it by an Expectation-Maximization(EM) algorithm with variational Bayes. Experimental results show that our method performs well on both synthetic and real data. Further, our method can predict overlapped block structure of a graphical model in general exponential families.

Cite this Paper

BibTeX
@InProceedings{pmlr-v38-sun15, title = {{Inferring Block Structure of Graphical Models in Exponential Families}}, author = {Sun, Siqi and Wang, Hai and Xu, Jinbo}, booktitle = {Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics}, pages = {939--947}, year = {2015}, editor = {Lebanon, Guy and Vishwanathan, S. V. N.}, volume = {38}, series = {Proceedings of Machine Learning Research}, address = {San Diego, California, USA}, month = {09--12 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v38/sun15.pdf}, url = {https://proceedings.mlr.press/v38/sun15.html}, abstract = {Learning the structure of a graphical model is a fundamental problem and it is used extensively to infer the relationship between random variables. In many real world applications, we usually have some prior knowledge about the underlying graph structure, such as degree distribution and block structure. In this paper, we propose a novel generative model for describing the block structure in general exponential families, and optimize it by an Expectation-Maximization(EM) algorithm with variational Bayes. Experimental results show that our method performs well on both synthetic and real data. Further, our method can predict overlapped block structure of a graphical model in general exponential families.} }
Endnote
%0 Conference Paper %T Inferring Block Structure of Graphical Models in Exponential Families %A Siqi Sun %A Hai Wang %A Jinbo Xu %B Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2015 %E Guy Lebanon %E S. V. N. Vishwanathan %F pmlr-v38-sun15 %I PMLR %P 939--947 %U https://proceedings.mlr.press/v38/sun15.html %V 38 %X Learning the structure of a graphical model is a fundamental problem and it is used extensively to infer the relationship between random variables. In many real world applications, we usually have some prior knowledge about the underlying graph structure, such as degree distribution and block structure. In this paper, we propose a novel generative model for describing the block structure in general exponential families, and optimize it by an Expectation-Maximization(EM) algorithm with variational Bayes. Experimental results show that our method performs well on both synthetic and real data. Further, our method can predict overlapped block structure of a graphical model in general exponential families.
RIS
TY - CPAPER TI - Inferring Block Structure of Graphical Models in Exponential Families AU - Siqi Sun AU - Hai Wang AU - Jinbo Xu BT - Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics DA - 2015/02/21 ED - Guy Lebanon ED - S. V. N. Vishwanathan ID - pmlr-v38-sun15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 38 SP - 939 EP - 947 L1 - http://proceedings.mlr.press/v38/sun15.pdf UR - https://proceedings.mlr.press/v38/sun15.html AB - Learning the structure of a graphical model is a fundamental problem and it is used extensively to infer the relationship between random variables. In many real world applications, we usually have some prior knowledge about the underlying graph structure, such as degree distribution and block structure. In this paper, we propose a novel generative model for describing the block structure in general exponential families, and optimize it by an Expectation-Maximization(EM) algorithm with variational Bayes. Experimental results show that our method performs well on both synthetic and real data. Further, our method can predict overlapped block structure of a graphical model in general exponential families. ER -
APA
Sun, S., Wang, H. & Xu, J.. (2015). Inferring Block Structure of Graphical Models in Exponential Families. Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 38:939-947 Available from https://proceedings.mlr.press/v38/sun15.html.

Related Material