In this paper, we present an Expectation Maximization algorithm, called Expectation Maximization Parameter learning for HIerarchical Probabilistic Logic programs (EMPHIL) which learns the parameters of Hierarchical Probabilistic Logic Programs from data. The algorithm converts an arithmetic circuit into a Bayesian network and performs the belief propagation algorithm over the corresponding factor graph.
Veröffentlicht am:
17th International Conference of the Italian Association for Artificial Intelligence (AI*IA 2018)