TY - GEN

T1 - Structured priors for structure learning

AU - Mansinghka, V. K.

AU - Kemp, C.

AU - Tenenbaum, J. B.

AU - Griffiths, Thomas L.

PY - 2006/12/1

Y1 - 2006/12/1

N2 - Traditional approaches to Bayes net structure learning typically assume little regularity in graph structure other than sparseness. However, in many cases, we expect more systematicity: variables in real-world systems often group into classes that predict the kinds of probabilistic dependencies they participate in. Here we capture this form of prior knowledge in a hierarchical Bayesian framework, and exploit it to enable structure learning and type discovery from small datasets. Specifically, we present a nonparametric generative model for directed acyclic graphs as a prior for Bayes net structure learning. Our model assumes that variables come in one or more classes and that the prior probability of an edge existing between two variables is a function only of their classes. We derive an MCMC algorithm for simultaneous inference of the number of classes, the class assignments of variables, and the Bayes net structure over variables. For several realistic, sparse datasets, we show that the bias towards systematicity of connections provided by our model can yield more accurate learned networks than the traditional approach of using a uniform prior, and that the classes found by our model are appropriate.

AB - Traditional approaches to Bayes net structure learning typically assume little regularity in graph structure other than sparseness. However, in many cases, we expect more systematicity: variables in real-world systems often group into classes that predict the kinds of probabilistic dependencies they participate in. Here we capture this form of prior knowledge in a hierarchical Bayesian framework, and exploit it to enable structure learning and type discovery from small datasets. Specifically, we present a nonparametric generative model for directed acyclic graphs as a prior for Bayes net structure learning. Our model assumes that variables come in one or more classes and that the prior probability of an edge existing between two variables is a function only of their classes. We derive an MCMC algorithm for simultaneous inference of the number of classes, the class assignments of variables, and the Bayes net structure over variables. For several realistic, sparse datasets, we show that the bias towards systematicity of connections provided by our model can yield more accurate learned networks than the traditional approach of using a uniform prior, and that the classes found by our model are appropriate.

UR - http://www.scopus.com/inward/record.url?scp=79958774082&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79958774082&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:79958774082

SN - 0974903922

SN - 9780974903927

T3 - Proceedings of the 22nd Conference on Uncertainty in Artificial Intelligence, UAI 2006

SP - 324

EP - 331

BT - Proceedings of the 22nd Conference on Uncertainty in Artificial Intelligence, UAI 2006

T2 - 22nd Conference on Uncertainty in Artificial Intelligence, UAI 2006

Y2 - 13 July 2006 through 16 July 2006

ER -