Seminar on Statistical Relational AI

Instructor: Professor Guy Van den Broeck <>
University of Trento


This seminar studies statistical representations of relational data. This includes formalisms from statistical relational learning, such as Markov logic networks, and relational Bayesian networks. The seminar also covers various probabilistic extensions of the relational data model, refered to as probabilistic databases, and probabilistic extensions of logical reasoning tasks, such as weighted model counting and integration. We will discuss their probabilistic inference problems, theoretical complexities, parameter, and structure learning algorithms, and representation power. The seminar also covers more general probabilistic programming languages and systems, which allow for more expressive statistical modeling.


Partial list of papers to present