Learning Bayesian and Markov Networks with an Unreliable Oracle

ArXi:2603.09563v1 Announce Type: new We study constraint-based structure learning of Marko networks and Bayesian networks in the presence of an unreliable conditional independence oracle that makes at most a bounded number of errors. For Marko networks, we observe that a low maximum number of vertex-wise disjoint paths implies that the structure is uniquely identifiable even if the number of errors is (moderately) exponential in the number of vertices.