Polynomial-time algorithm for learning optimal tree-augmented dynamic Bayesian networks

The identification of conditional dependences in longitudinal data is provided through structure learning of dynamic Bayesian networks (DBN). Several methods for DBN learning are concerned with identifying inter-slice dependences, but often disregard the intra-slice connectivity. We propose an algorithm that jointly finds the optimal inter and intra time-slice connectivity in a transition network. The search space is constrained to a class of networks designated by tree-augmented DBN, leading to polynomial time complexity. We assess the effectiveness of the algorithm on simulated data and compare the results to those obtained by a state of the art DBN learning implementation, showing that the proposed algorithm performs very well throughout the different experiments. Further experimental validation is made on real data, by identifying non-stationary gene regulatory networks of Drosophila melanogaster.