Adaptive Variable Clustering in Gaussian Graphical Models

Siqi Sun, Yuancheng Zhu, Jinbo Xu
Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33:931-939, 2014.

Abstract

Gaussian graphical models (GGMs) are widely-used to describe the relationship between random variables. In many real-world applications, GGMs have a block structure in the sense that the variables can be clustered into groups so that inter-group correlation is much weaker than intra-group correlation. We present a novel nonparametric Bayesian generative model for such a block-structured GGM and an efficient inference algorithm to find the clustering of variables in this GGM by combining a Gibbs sampler and a split-merge Metropolis-Hastings algorithm. Experimental results show that our method performs well on both synthetic and real data. In particular, our method outperforms generic clustering algorithms and can automatically identify the true number of clusters.

Cite this Paper

BibTeX
@InProceedings{pmlr-v33-sun14, title = {{ Adaptive Variable Clustering in Gaussian Graphical Models}}, author = {Sun, Siqi and Zhu, Yuancheng and Xu, Jinbo}, booktitle = {Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics}, pages = {931--939}, year = {2014}, editor = {Kaski, Samuel and Corander, Jukka}, volume = {33}, series = {Proceedings of Machine Learning Research}, address = {Reykjavik, Iceland}, month = {22--25 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v33/sun14.pdf}, url = {https://proceedings.mlr.press/v33/sun14.html}, abstract = {Gaussian graphical models (GGMs) are widely-used to describe the relationship between random variables. In many real-world applications, GGMs have a block structure in the sense that the variables can be clustered into groups so that inter-group correlation is much weaker than intra-group correlation. We present a novel nonparametric Bayesian generative model for such a block-structured GGM and an efficient inference algorithm to find the clustering of variables in this GGM by combining a Gibbs sampler and a split-merge Metropolis-Hastings algorithm. Experimental results show that our method performs well on both synthetic and real data. In particular, our method outperforms generic clustering algorithms and can automatically identify the true number of clusters.} }
Endnote
%0 Conference Paper %T Adaptive Variable Clustering in Gaussian Graphical Models %A Siqi Sun %A Yuancheng Zhu %A Jinbo Xu %B Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2014 %E Samuel Kaski %E Jukka Corander %F pmlr-v33-sun14 %I PMLR %P 931--939 %U https://proceedings.mlr.press/v33/sun14.html %V 33 %X Gaussian graphical models (GGMs) are widely-used to describe the relationship between random variables. In many real-world applications, GGMs have a block structure in the sense that the variables can be clustered into groups so that inter-group correlation is much weaker than intra-group correlation. We present a novel nonparametric Bayesian generative model for such a block-structured GGM and an efficient inference algorithm to find the clustering of variables in this GGM by combining a Gibbs sampler and a split-merge Metropolis-Hastings algorithm. Experimental results show that our method performs well on both synthetic and real data. In particular, our method outperforms generic clustering algorithms and can automatically identify the true number of clusters.
RIS
TY - CPAPER TI - Adaptive Variable Clustering in Gaussian Graphical Models AU - Siqi Sun AU - Yuancheng Zhu AU - Jinbo Xu BT - Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics DA - 2014/04/02 ED - Samuel Kaski ED - Jukka Corander ID - pmlr-v33-sun14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 33 SP - 931 EP - 939 L1 - http://proceedings.mlr.press/v33/sun14.pdf UR - https://proceedings.mlr.press/v33/sun14.html AB - Gaussian graphical models (GGMs) are widely-used to describe the relationship between random variables. In many real-world applications, GGMs have a block structure in the sense that the variables can be clustered into groups so that inter-group correlation is much weaker than intra-group correlation. We present a novel nonparametric Bayesian generative model for such a block-structured GGM and an efficient inference algorithm to find the clustering of variables in this GGM by combining a Gibbs sampler and a split-merge Metropolis-Hastings algorithm. Experimental results show that our method performs well on both synthetic and real data. In particular, our method outperforms generic clustering algorithms and can automatically identify the true number of clusters. ER -
APA
Sun, S., Zhu, Y. & Xu, J.. (2014). Adaptive Variable Clustering in Gaussian Graphical Models. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 33:931-939 Available from https://proceedings.mlr.press/v33/sun14.html.

Related Material