Discovering the Hidden Structure of Complex Dynamic Systems

Dynamic Bayesian networks provide a compact and natural representation for complex dynamic systems. However, in many cases, there is no expert available from whom a model can be elicited. Learning provides an alternative approach for constructing models of dynamic systems. In this paper, we address some of the crucial computational aspects of learning the structure of dynamic systems, particularly those where some relevant variables are partially observed or even entirely unknown. Our approach is based on the Structural Expectation Maximization (SEM) algorithm. The main computational cost of the SEM algorithm is the gathering of expected sufficient statistics. We propose a novel approximation scheme that allows these sufficient statistics to be computed efficiently. We also investigate the fundamental problem of discovering the existence of hidden variables without exhaustive and expensive search. Our approach is based on the observation that, in dynamic systems, ignoring a hidden variable typically results in a violation of the Markov property. Thus, our algorithm searches for such violations in the data, and introduces hidden variables to explain them. We provide empirical results showing that the algorithm is able to learn the dynamics of complex systems in a computationally tractable way.

[1]  David J. Spiegelhalter,et al.  Local computations with probabilities on graphical structures and their application to expert systems , 1990 .

[2]  Wray L. Buntine.  Theory Refinement on Bayesian Networks , 1991, UAI.

[3]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[4]  David Heckerman,et al.  A Tutorial on Learning with Bayesian Networks , 1998, Learning in Graphical Models.

[5]  S. Lauritzen.  The EM algorithm for graphical association models with missing data , 1995 .

[6]  Xavier Boyen,et al.  Exploiting the Architecture of Dynamic Systems , 1999, AAAI/IAAI.

[7]  Stuart J. Russell,et al.  The BATmobile: Towards a Bayesian Automated Taxi , 1995, IJCAI.

[8]  P. Spirtes,et al.  Causation, prediction, and search , 1993 .

[9]  Biing-Hwang Juang,et al.  Mixture autoregressive hidden Markov models for speech signals , 1985, IEEE Trans. Acoust. Speech Signal Process..

[10]  Wray L. BuntineRIACS.  Theory Reenement on Bayesian Networks , 1991 .

[11]  Kevin P. Murphy,et al.  Learning the Structure of Dynamic Probabilistic Networks , 1998, UAI.

[12]  Nir Friedman,et al.  Learning Belief Networks in the Presence of Missing Values and Hidden Variables , 1997, ICML.

[13]  David Heckerman,et al.  Learning Bayesian Networks: Search Methods and Experimental Results , 1995 .

[14]  Paul Dagum,et al.  Forecasting Sleep Apnea with Dynamic Network Models , 1993, UAI.

[15]  Xavier Boyen,et al.  Approximate Learning of Dynamic Models , 1998, NIPS.

[16]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[17]  Nir Friedman,et al.  The Bayesian Structural EM Algorithm , 1998, UAI.

[18]  Xavier Boyen,et al.  Tractable Inference for Complex Stochastic Processes , 1998, UAI.

[19]  Nir Friedman,et al.  Learning Bayesian Network Structure from Massive Datasets: The "Sparse Candidate" Algorithm , 1999, UAI.

[20]  Prakash P. Shenoy,et al.  Axioms for probability and belief-function proagation , 1990, UAI.