The Constraint Reasoning and Optimization contributes to PGM 2018, an international conference on focusing on different aspects of probabilistic graphical models, with two research papers.
Learning Optimal Causal Graphs with Exact Search by Kari Rantanen, Antti Hyttinen and Matti Järvisalo, presents the first truly specialized exact search algorithm for optimal causal graphs in a general model space, allowing both cycles and latent confounding variables, and combining problem-specific branch-and-bound style search with linear programming based bounding. The resulting system bcause constitutes empirically a new state-of-the-art approach to the problem.
Structure Learning for Bayesian Networks over Labeled DAGs co-authored by Antti Hyttinen together with Johan Pensar, Juha Kontinen, and Jukka Corander, presents a constraint-based LPC algorithm and a score-based method for learning graphical models representing context specific independencies.