Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Lehr-und Forschungsgebiet Informatik 5 Zeit: Donnerstag, 12. Juni 2008, 16:30 Uhr

Summary: Lehr- und Forschungsgebiet Informatik 5
Zeit: Donnerstag, 12. Juni 2008, 16:30 Uhr
Ort: AH I, Ahornstr. 55
Referent: Prof. Luc De Raedt
Katholieke Universiteit Leuven
Titel: An Introduction to Statistical Relational Learning:
A Logical Perspective
Statistical Relational Learning is a new subfield of artificial intelligence lying at the
intersection of machine learning, reasoning about undertainty and relational and
logical representations. It aims at developing models that can elegantly deal with
objects as well as the relationships that hold amongst them. In this talk, I shall first
motivate this research stream using a number of applications and then analyse its
state-of-the-art taking a logical perspective. More specifically, I shall explore the
relationships between the traditional probabilistic models, which work essentially with
a propositional representation, and their upgrades within statistical relational learning
and introduce these using a number of examples. The examples used will include the
upgrading of Bayesian or Markov networks (towards PRMs, BLPs or MLNs) and
those of probabilistic Context Free Grammars and HMMs (towards SLPs, PRISM,


Source: Ábrahám, Erika - Fachgruppe Informatik, Rheinisch Westfälische Technische Hochschule Aachen (RWTH)


Collections: Computer Technologies and Information Sciences