Giuseppe Marra (DIISM, University of Siena)
Jun 7, 2018 – 9:30 AM
DIISM, Artificial Intelligence laboratory (room 201), Siena SI
A fundamental challenge in developing high-impact machine learning technologies is balancing the need to model rich, structured domains with the ability to scale to big data. Many important problem areas are both richly structured and large scale, from social and biological networks, to knowledge graphs and the Web, to images, video, and natural language. In this seminar, we present PSL, Probabilistic Soft Logic , a probabilistic programming language that allows an easy definition of probabilistic graphical models using a first-order logic based syntax. We introduce Hinge-Loss Markov Random Fields, a probabilistic graphical model that shows that performing inference in a Lukasiewicz interpretation of the logical problem defined by PSL is equivalent to solving a relaxed max-SAT problem.
 Hinge-loss Markov Random Fields and Probabilistic Soft Logic
Stephen H. Bach, Matthias Broecheler, Bert Huang, and Lise Getoor