Joint Structure Learning of Multiple Non-Exchangeable Networks

Chris Oates, Sach Mukherjee
Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33:687-695, 2014.

Abstract

Several methods have recently been developed for joint structure learning of multiple (related) graphical models or networks. These methods treat individual networks as exchangeable, such that each pair of networks are equally encouraged to have similar structures. However, in many practical applications, exchangeability in this sense does not hold, as some pairs of networks may be more closely related than others, for example due to group and sub-group structures in the data. Here we present a novel Bayesian formulation that generalises joint structure learning beyond the exchangeable case. Moreover (i) a novel default prior over the joint structure space is proposed that requires no user input; (ii) latent networks are permitted; (iii) for time series data and dynamic Bayesian networks, an efficient, exact algorithm is provided. We present empirical results on non-exchangeable populations, including a real example from cancer biology, where cell-line specific networks are related according to known genomic features.

Cite this Paper

BibTeX
@InProceedings{pmlr-v33-oates14, title = {{Joint Structure Learning of Multiple Non-Exchangeable Networks}}, author = {Oates, Chris and Mukherjee, Sach}, booktitle = {Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics}, pages = {687--695}, 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/oates14.pdf}, url = {https://proceedings.mlr.press/v33/oates14.html}, abstract = {Several methods have recently been developed for joint structure learning of multiple (related) graphical models or networks. These methods treat individual networks as exchangeable, such that each pair of networks are equally encouraged to have similar structures. However, in many practical applications, exchangeability in this sense does not hold, as some pairs of networks may be more closely related than others, for example due to group and sub-group structures in the data. Here we present a novel Bayesian formulation that generalises joint structure learning beyond the exchangeable case. Moreover (i) a novel default prior over the joint structure space is proposed that requires no user input; (ii) latent networks are permitted; (iii) for time series data and dynamic Bayesian networks, an efficient, exact algorithm is provided. We present empirical results on non-exchangeable populations, including a real example from cancer biology, where cell-line specific networks are related according to known genomic features.} }
Endnote
%0 Conference Paper %T Joint Structure Learning of Multiple Non-Exchangeable Networks %A Chris Oates %A Sach Mukherjee %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-oates14 %I PMLR %P 687--695 %U https://proceedings.mlr.press/v33/oates14.html %V 33 %X Several methods have recently been developed for joint structure learning of multiple (related) graphical models or networks. These methods treat individual networks as exchangeable, such that each pair of networks are equally encouraged to have similar structures. However, in many practical applications, exchangeability in this sense does not hold, as some pairs of networks may be more closely related than others, for example due to group and sub-group structures in the data. Here we present a novel Bayesian formulation that generalises joint structure learning beyond the exchangeable case. Moreover (i) a novel default prior over the joint structure space is proposed that requires no user input; (ii) latent networks are permitted; (iii) for time series data and dynamic Bayesian networks, an efficient, exact algorithm is provided. We present empirical results on non-exchangeable populations, including a real example from cancer biology, where cell-line specific networks are related according to known genomic features.
RIS
TY - CPAPER TI - Joint Structure Learning of Multiple Non-Exchangeable Networks AU - Chris Oates AU - Sach Mukherjee 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-oates14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 33 SP - 687 EP - 695 L1 - http://proceedings.mlr.press/v33/oates14.pdf UR - https://proceedings.mlr.press/v33/oates14.html AB - Several methods have recently been developed for joint structure learning of multiple (related) graphical models or networks. These methods treat individual networks as exchangeable, such that each pair of networks are equally encouraged to have similar structures. However, in many practical applications, exchangeability in this sense does not hold, as some pairs of networks may be more closely related than others, for example due to group and sub-group structures in the data. Here we present a novel Bayesian formulation that generalises joint structure learning beyond the exchangeable case. Moreover (i) a novel default prior over the joint structure space is proposed that requires no user input; (ii) latent networks are permitted; (iii) for time series data and dynamic Bayesian networks, an efficient, exact algorithm is provided. We present empirical results on non-exchangeable populations, including a real example from cancer biology, where cell-line specific networks are related according to known genomic features. ER -
APA
Oates, C. & Mukherjee, S.. (2014). Joint Structure Learning of Multiple Non-Exchangeable Networks. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 33:687-695 Available from https://proceedings.mlr.press/v33/oates14.html.

Related Material