Discriminative Structure and Parameter Learning for Markov Logic Networks

Markov logic networks (MLNs) are an expressive representation for statistical relational learning that generalizes both first-order logic and graphical models. Existing methods for learning the logica