Bayesian regularization of non-homogeneous dynamic Bayesian networks by globally coupling interaction parameters

Marco Grzegorzyk, Dirk Husmeier
Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, PMLR 22:467-476, 2012.

Abstract

To relax the homogeneity assumption of classical dynamic Bayesian networks (DBNs), various recent studies have combined DBNs with multiple changepoint processes. The underlying assumption is that the parameters associated with time series segments delimited by multiple changepoints are a priori independent. Under weak regularity conditions, the parameters can be integrated out in the likelihood, leading to a closed-form expression of the marginal likelihood. However, the assumption of prior independence is unrealistic in many real-world applications, where the segment-specific regulatory relationships among the interdependent quantities tend to undergo gradual evolutionary adaptations. We therefore propose a Bayesian coupling scheme to introduce systematic information sharing among the segment-specific interaction parameters. We investigate the effect this model improvement has on the network reconstruction accuracy in a reverse engineering context, where the objective is to learn the structure of a gene regulatory network from temporal gene expression profiles.

Cite this Paper

BibTeX
@InProceedings{pmlr-v22-grzegorzyk12, title = {Bayesian regularization of non-homogeneous dynamic Bayesian networks by globally coupling interaction parameters}, author = {Grzegorzyk, Marco and Husmeier, Dirk}, booktitle = {Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics}, pages = {467--476}, year = {2012}, editor = {Lawrence, Neil D. and Girolami, Mark}, volume = {22}, series = {Proceedings of Machine Learning Research}, address = {La Palma, Canary Islands}, month = {21--23 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v22/grzegorzyk12/grzegorzyk12.pdf}, url = {https://proceedings.mlr.press/v22/grzegorzyk12.html}, abstract = {To relax the homogeneity assumption of classical dynamic Bayesian networks (DBNs), various recent studies have combined DBNs with multiple changepoint processes. The underlying assumption is that the parameters associated with time series segments delimited by multiple changepoints are a priori independent. Under weak regularity conditions, the parameters can be integrated out in the likelihood, leading to a closed-form expression of the marginal likelihood. However, the assumption of prior independence is unrealistic in many real-world applications, where the segment-specific regulatory relationships among the interdependent quantities tend to undergo gradual evolutionary adaptations. We therefore propose a Bayesian coupling scheme to introduce systematic information sharing among the segment-specific interaction parameters. We investigate the effect this model improvement has on the network reconstruction accuracy in a reverse engineering context, where the objective is to learn the structure of a gene regulatory network from temporal gene expression profiles.} }
Endnote
%0 Conference Paper %T Bayesian regularization of non-homogeneous dynamic Bayesian networks by globally coupling interaction parameters %A Marco Grzegorzyk %A Dirk Husmeier %B Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2012 %E Neil D. Lawrence %E Mark Girolami %F pmlr-v22-grzegorzyk12 %I PMLR %P 467--476 %U https://proceedings.mlr.press/v22/grzegorzyk12.html %V 22 %X To relax the homogeneity assumption of classical dynamic Bayesian networks (DBNs), various recent studies have combined DBNs with multiple changepoint processes. The underlying assumption is that the parameters associated with time series segments delimited by multiple changepoints are a priori independent. Under weak regularity conditions, the parameters can be integrated out in the likelihood, leading to a closed-form expression of the marginal likelihood. However, the assumption of prior independence is unrealistic in many real-world applications, where the segment-specific regulatory relationships among the interdependent quantities tend to undergo gradual evolutionary adaptations. We therefore propose a Bayesian coupling scheme to introduce systematic information sharing among the segment-specific interaction parameters. We investigate the effect this model improvement has on the network reconstruction accuracy in a reverse engineering context, where the objective is to learn the structure of a gene regulatory network from temporal gene expression profiles.
RIS
TY - CPAPER TI - Bayesian regularization of non-homogeneous dynamic Bayesian networks by globally coupling interaction parameters AU - Marco Grzegorzyk AU - Dirk Husmeier BT - Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics DA - 2012/03/21 ED - Neil D. Lawrence ED - Mark Girolami ID - pmlr-v22-grzegorzyk12 PB - PMLR DP - Proceedings of Machine Learning Research VL - 22 SP - 467 EP - 476 L1 - http://proceedings.mlr.press/v22/grzegorzyk12/grzegorzyk12.pdf UR - https://proceedings.mlr.press/v22/grzegorzyk12.html AB - To relax the homogeneity assumption of classical dynamic Bayesian networks (DBNs), various recent studies have combined DBNs with multiple changepoint processes. The underlying assumption is that the parameters associated with time series segments delimited by multiple changepoints are a priori independent. Under weak regularity conditions, the parameters can be integrated out in the likelihood, leading to a closed-form expression of the marginal likelihood. However, the assumption of prior independence is unrealistic in many real-world applications, where the segment-specific regulatory relationships among the interdependent quantities tend to undergo gradual evolutionary adaptations. We therefore propose a Bayesian coupling scheme to introduce systematic information sharing among the segment-specific interaction parameters. We investigate the effect this model improvement has on the network reconstruction accuracy in a reverse engineering context, where the objective is to learn the structure of a gene regulatory network from temporal gene expression profiles. ER -
APA
Grzegorzyk, M. & Husmeier, D.. (2012). Bayesian regularization of non-homogeneous dynamic Bayesian networks by globally coupling interaction parameters. Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 22:467-476 Available from https://proceedings.mlr.press/v22/grzegorzyk12.html.

Related Material