Effective Learning of Probabilistic Models from Structured Data

Abstract: Statistical Relational Learning (SRL) Models combine the powerful formalisms of probability theory and first-order logic to handle uncertainty in large, complex problems. In this talk, I will review the progress in this area. First, I will provide a historical outlook of this field by covering some of the seminal work in the area. Then, I will discuss state-of-the-art learning method in this area that is representation independent. I will also discuss about how this learning can leverage from rich human interaction. Finally I will conclude by discussing some of the possible “killer apps” for this area.