An Improved Structural EM to Learn Dynamic Bayesian Nets

This paper addresses the problem of learning structure and parameters of Bayesian and Dynamic Bayesian networks from data based on the Bayesian Information Criterion. We describe a procedure to map the problem of the dynamic case into a corre-sponding augmented Bayesian network through the use of structural constraints. Because the algo-rithm is exact and anytime, it is well suitable for a structural Expectation–Maximization (EM) method where the only source of approximation is due to the EM itself. We show empirically that the use a global maximizer inside the structural EM is computation-ally feasible and leads to more accurate models.