Dynamic Copula Networks for Modeling Real-valued Time Series

Elad Eban, Gideon Rothschild, Adi Mizrahi, Israel Nelken, Gal Elidan
Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics, PMLR 31:247-255, 2013.

Abstract

Probabilistic modeling of temporal phenomena is of central importance in a variety of fields ranging from neuroscience to economics to speech recognition. While the task has received extensive attention in recent decades, learning temporal models for multivariate real-valued data that is non-Gaussian is still a formidable challenge. Recently, the power of copulas, a framework for representing complex multi-modal and heavy-tailed distributions, was fused with the formalism of Bayesian networks to allow for flexible modeling of high-dimensional distributions. In this work we introduce Dynamic Copula Bayesian Networks, a generalization aimed at capturing the distribution of rich temporal sequences. We apply our model to three markedly different real-life domains and demonstrate substantial quantitative and qualitative advantage.

Cite this Paper

BibTeX
@InProceedings{pmlr-v31-eban13a, title = {Dynamic Copula Networks for Modeling Real-valued Time Series}, author = {Eban, Elad and Rothschild, Gideon and Mizrahi, Adi and Nelken, Israel and Elidan, Gal}, booktitle = {Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics}, pages = {247--255}, year = {2013}, editor = {Carvalho, Carlos M. and Ravikumar, Pradeep}, volume = {31}, series = {Proceedings of Machine Learning Research}, address = {Scottsdale, Arizona, USA}, month = {29 Apr--01 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v31/eban13a.pdf}, url = {https://proceedings.mlr.press/v31/eban13a.html}, abstract = {Probabilistic modeling of temporal phenomena is of central importance in a variety of fields ranging from neuroscience to economics to speech recognition. While the task has received extensive attention in recent decades, learning temporal models for multivariate real-valued data that is non-Gaussian is still a formidable challenge. Recently, the power of copulas, a framework for representing complex multi-modal and heavy-tailed distributions, was fused with the formalism of Bayesian networks to allow for flexible modeling of high-dimensional distributions. In this work we introduce Dynamic Copula Bayesian Networks, a generalization aimed at capturing the distribution of rich temporal sequences. We apply our model to three markedly different real-life domains and demonstrate substantial quantitative and qualitative advantage.} }
Endnote
%0 Conference Paper %T Dynamic Copula Networks for Modeling Real-valued Time Series %A Elad Eban %A Gideon Rothschild %A Adi Mizrahi %A Israel Nelken %A Gal Elidan %B Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2013 %E Carlos M. Carvalho %E Pradeep Ravikumar %F pmlr-v31-eban13a %I PMLR %P 247--255 %U https://proceedings.mlr.press/v31/eban13a.html %V 31 %X Probabilistic modeling of temporal phenomena is of central importance in a variety of fields ranging from neuroscience to economics to speech recognition. While the task has received extensive attention in recent decades, learning temporal models for multivariate real-valued data that is non-Gaussian is still a formidable challenge. Recently, the power of copulas, a framework for representing complex multi-modal and heavy-tailed distributions, was fused with the formalism of Bayesian networks to allow for flexible modeling of high-dimensional distributions. In this work we introduce Dynamic Copula Bayesian Networks, a generalization aimed at capturing the distribution of rich temporal sequences. We apply our model to three markedly different real-life domains and demonstrate substantial quantitative and qualitative advantage.
RIS
TY - CPAPER TI - Dynamic Copula Networks for Modeling Real-valued Time Series AU - Elad Eban AU - Gideon Rothschild AU - Adi Mizrahi AU - Israel Nelken AU - Gal Elidan BT - Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics DA - 2013/04/29 ED - Carlos M. Carvalho ED - Pradeep Ravikumar ID - pmlr-v31-eban13a PB - PMLR DP - Proceedings of Machine Learning Research VL - 31 SP - 247 EP - 255 L1 - http://proceedings.mlr.press/v31/eban13a.pdf UR - https://proceedings.mlr.press/v31/eban13a.html AB - Probabilistic modeling of temporal phenomena is of central importance in a variety of fields ranging from neuroscience to economics to speech recognition. While the task has received extensive attention in recent decades, learning temporal models for multivariate real-valued data that is non-Gaussian is still a formidable challenge. Recently, the power of copulas, a framework for representing complex multi-modal and heavy-tailed distributions, was fused with the formalism of Bayesian networks to allow for flexible modeling of high-dimensional distributions. In this work we introduce Dynamic Copula Bayesian Networks, a generalization aimed at capturing the distribution of rich temporal sequences. We apply our model to three markedly different real-life domains and demonstrate substantial quantitative and qualitative advantage. ER -
APA
Eban, E., Rothschild, G., Mizrahi, A., Nelken, I. & Elidan, G.. (2013). Dynamic Copula Networks for Modeling Real-valued Time Series. Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 31:247-255 Available from https://proceedings.mlr.press/v31/eban13a.html.

Related Material