Recovering Networks from Distance Data

Sandhya Prabhakaran, Karin J. Metzner, Alexander Böhm, Volker Roth
Proceedings of the Asian Conference on Machine Learning, PMLR 25:349-364, 2012.

Abstract

A fully probabilistic approach to reconstructing Gaussian graphical models from distance data is presented. The main idea is to extend the usual central Wishart model in traditional methods to using a likelihood depending only on pairwise distances, thus being independent of geometric assumptions about the underlying Euclidean space. This extension has two advantages: the model becomes invariant against potential bias terms in the measurements, and can be used in situations which on input use a kernel- or distance matrix, without requiring direct access to the underlying vectors. The latter aspect opens up a huge new application field for Gaussian graphical models, as network reconstruction is now possible from any Mercer kernel, be it on graphs, strings, probabilities or more complex objects. We combine this likelihood with a suitable prior to enable Bayesian network inference. We present an efficient MCMC sampler for this model and discuss the estimation of module networks. Experiments depict the high quality and usefulness of the inferred networks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v25-prabhakaran12, title = {Recovering Networks from Distance Data}, author = {Prabhakaran, Sandhya and Metzner, Karin J. and Böhm, Alexander and Roth, Volker}, booktitle = {Proceedings of the Asian Conference on Machine Learning}, pages = {349--364}, year = {2012}, editor = {Hoi, Steven C. H. and Buntine, Wray}, volume = {25}, series = {Proceedings of Machine Learning Research}, address = {Singapore Management University, Singapore}, month = {04--06 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v25/prabhakaran12/prabhakaran12.pdf}, url = {https://proceedings.mlr.press/v25/prabhakaran12.html}, abstract = {A fully probabilistic approach to reconstructing Gaussian graphical models from distance data is presented. The main idea is to extend the usual central Wishart model in traditional methods to using a likelihood depending only on pairwise distances, thus being independent of geometric assumptions about the underlying Euclidean space. This extension has two advantages: the model becomes invariant against potential bias terms in the measurements, and can be used in situations which on input use a kernel- or distance matrix, without requiring direct access to the underlying vectors. The latter aspect opens up a huge new application field for Gaussian graphical models, as network reconstruction is now possible from any Mercer kernel, be it on graphs, strings, probabilities or more complex objects. We combine this likelihood with a suitable prior to enable Bayesian network inference. We present an efficient MCMC sampler for this model and discuss the estimation of module networks. Experiments depict the high quality and usefulness of the inferred networks.} }
Endnote
%0 Conference Paper %T Recovering Networks from Distance Data %A Sandhya Prabhakaran %A Karin J. Metzner %A Alexander Böhm %A Volker Roth %B Proceedings of the Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2012 %E Steven C. H. Hoi %E Wray Buntine %F pmlr-v25-prabhakaran12 %I PMLR %P 349--364 %U https://proceedings.mlr.press/v25/prabhakaran12.html %V 25 %X A fully probabilistic approach to reconstructing Gaussian graphical models from distance data is presented. The main idea is to extend the usual central Wishart model in traditional methods to using a likelihood depending only on pairwise distances, thus being independent of geometric assumptions about the underlying Euclidean space. This extension has two advantages: the model becomes invariant against potential bias terms in the measurements, and can be used in situations which on input use a kernel- or distance matrix, without requiring direct access to the underlying vectors. The latter aspect opens up a huge new application field for Gaussian graphical models, as network reconstruction is now possible from any Mercer kernel, be it on graphs, strings, probabilities or more complex objects. We combine this likelihood with a suitable prior to enable Bayesian network inference. We present an efficient MCMC sampler for this model and discuss the estimation of module networks. Experiments depict the high quality and usefulness of the inferred networks.
RIS
TY - CPAPER TI - Recovering Networks from Distance Data AU - Sandhya Prabhakaran AU - Karin J. Metzner AU - Alexander Böhm AU - Volker Roth BT - Proceedings of the Asian Conference on Machine Learning DA - 2012/11/17 ED - Steven C. H. Hoi ED - Wray Buntine ID - pmlr-v25-prabhakaran12 PB - PMLR DP - Proceedings of Machine Learning Research VL - 25 SP - 349 EP - 364 L1 - http://proceedings.mlr.press/v25/prabhakaran12/prabhakaran12.pdf UR - https://proceedings.mlr.press/v25/prabhakaran12.html AB - A fully probabilistic approach to reconstructing Gaussian graphical models from distance data is presented. The main idea is to extend the usual central Wishart model in traditional methods to using a likelihood depending only on pairwise distances, thus being independent of geometric assumptions about the underlying Euclidean space. This extension has two advantages: the model becomes invariant against potential bias terms in the measurements, and can be used in situations which on input use a kernel- or distance matrix, without requiring direct access to the underlying vectors. The latter aspect opens up a huge new application field for Gaussian graphical models, as network reconstruction is now possible from any Mercer kernel, be it on graphs, strings, probabilities or more complex objects. We combine this likelihood with a suitable prior to enable Bayesian network inference. We present an efficient MCMC sampler for this model and discuss the estimation of module networks. Experiments depict the high quality and usefulness of the inferred networks. ER -
APA
Prabhakaran, S., Metzner, K.J., Böhm, A. & Roth, V.. (2012). Recovering Networks from Distance Data. Proceedings of the Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 25:349-364 Available from https://proceedings.mlr.press/v25/prabhakaran12.html.

Related Material