Markov Logic: A Unifying Framework for Statistical Relational Learning

Pedro Domingos and Matthew Richardson

Abstract:

Interest in statistical relational learning (SRL) has grown rapidly in recent years. Several key SRL tasks have been identified, and a large number of approaches have been proposed. Increasingly, a unifying framework is needed to facilitate transfer of knowledge across tasks and approaches, to compare approaches, and to help bring structure to the field. We propose Markov logic as such a framework. Syntactically, Markov logic is indistinguishable from first-order logic, except that each formula has a weight attached. Semantically, a set of Markov logic formulas represents a probability distribution over possible worlds, in the form of a log-linear model with one feature per grounding of a formula in the set, with the corresponding weight. We show how approaches like probabilistic relational models, knowledge- based model construction and stochastic logic programs can be mapped into Markov logic. We also show how tasks like collective classification, link prediction, link- based clustering, social network modeling, and object identification can be concisely formulated in Markov logic. Finally, we develop learning and inference algorithms for Markov logic, and report experimental results on a link prediction task.

Download:

Paper (PDF)