Curriculum Learning of Bayesian Network Structures

Bayesian networks (BNs) are directed graphical models that have been widely used in various tasks for probabilistic reasoning and causal modeling. One major challenge in these tasks is to learn the BN structures from data. In this paper, we propose a novel heuristic algorithm for BN structure learning that takes advantage of the idea of curriculum learning. Our algorithm learns the BN structure by stages. At each stage a subnet is learned over a selected subset of the random variables conditioned on fixed values of the rest of the variables. The selected subset grows with stages and eventually includes all the variables. We prove theoretical advantages of our algorithm and also empirically show that it outperformed the state-of-the-art heuristic approach in learning BN structures.