Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations

Aaron Schein, Mingyuan Zhou, David Blei, Hanna Wallach
Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:2810-2819, 2016.

Abstract

We introduce Bayesian Poisson Tucker decomposition (BPTD) for modeling country–country interaction event data. These data consist of interaction events of the form “country i took action a toward country j at time t.” BPTD discovers overlapping country–community memberships, including the number of latent communities. In addition, it discovers directed community–community interaction networks that are specific to “topics” of action types and temporal “regimes.” We show that BPTD yields an efficient MCMC inference algorithm and achieves better predictive performance than related models. We also demonstrate that it discovers interpretable latent structure that agrees with our knowledge of international relations.

Cite this Paper


BibTeX
@InProceedings{pmlr-v48-schein16, title = {Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations}, author = {Schein, Aaron and Zhou, Mingyuan and Blei, David and Wallach, Hanna}, booktitle = {Proceedings of The 33rd International Conference on Machine Learning}, pages = {2810--2819}, year = {2016}, editor = {Balcan, Maria Florina and Weinberger, Kilian Q.}, volume = {48}, series = {Proceedings of Machine Learning Research}, address = {New York, New York, USA}, month = {20--22 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v48/schein16.pdf}, url = {http://proceedings.mlr.press/v48/schein16.html}, abstract = {We introduce Bayesian Poisson Tucker decomposition (BPTD) for modeling country–country interaction event data. These data consist of interaction events of the form “country i took action a toward country j at time t.” BPTD discovers overlapping country–community memberships, including the number of latent communities. In addition, it discovers directed community–community interaction networks that are specific to “topics” of action types and temporal “regimes.” We show that BPTD yields an efficient MCMC inference algorithm and achieves better predictive performance than related models. We also demonstrate that it discovers interpretable latent structure that agrees with our knowledge of international relations.} }
Endnote
%0 Conference Paper %T Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations %A Aaron Schein %A Mingyuan Zhou %A David Blei %A Hanna Wallach %B Proceedings of The 33rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Maria Florina Balcan %E Kilian Q. Weinberger %F pmlr-v48-schein16 %I PMLR %P 2810--2819 %U http://proceedings.mlr.press/v48/schein16.html %V 48 %X We introduce Bayesian Poisson Tucker decomposition (BPTD) for modeling country–country interaction event data. These data consist of interaction events of the form “country i took action a toward country j at time t.” BPTD discovers overlapping country–community memberships, including the number of latent communities. In addition, it discovers directed community–community interaction networks that are specific to “topics” of action types and temporal “regimes.” We show that BPTD yields an efficient MCMC inference algorithm and achieves better predictive performance than related models. We also demonstrate that it discovers interpretable latent structure that agrees with our knowledge of international relations.
RIS
TY - CPAPER TI - Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations AU - Aaron Schein AU - Mingyuan Zhou AU - David Blei AU - Hanna Wallach BT - Proceedings of The 33rd International Conference on Machine Learning DA - 2016/06/11 ED - Maria Florina Balcan ED - Kilian Q. Weinberger ID - pmlr-v48-schein16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 48 SP - 2810 EP - 2819 L1 - http://proceedings.mlr.press/v48/schein16.pdf UR - http://proceedings.mlr.press/v48/schein16.html AB - We introduce Bayesian Poisson Tucker decomposition (BPTD) for modeling country–country interaction event data. These data consist of interaction events of the form “country i took action a toward country j at time t.” BPTD discovers overlapping country–community memberships, including the number of latent communities. In addition, it discovers directed community–community interaction networks that are specific to “topics” of action types and temporal “regimes.” We show that BPTD yields an efficient MCMC inference algorithm and achieves better predictive performance than related models. We also demonstrate that it discovers interpretable latent structure that agrees with our knowledge of international relations. ER -
APA
Schein, A., Zhou, M., Blei, D. & Wallach, H.. (2016). Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations. Proceedings of The 33rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 48:2810-2819 Available from http://proceedings.mlr.press/v48/schein16.html.

Related Material