"2006 UPF Course on Principles
of AI Problem Solving"
Universitat Pompeu Fabra, March and April
2006
Coordinator: Hector Geffner
hector.geffner at upf.edu
This edition of the course will be given by the following two leading
researchers: Rina Dechter from the
University of California at Irvine (UCI), and Andrew Barto from the
University of Massachusetts at Amherst (UMass). The two
parts of the course are:
Inference and Search in
Graphical Models (Constraint and Probabilistic Bayesian Networks)
Rina Dechter
March 27-31, 2006
10h-13h
Room 102
Slides: Introduction,
Constraint
Networks (Inference), Constraint
Networks (Search), Bayesian Networks
1, Bayesian
Networks 2, Constraint
Optimization
A paper that summarizes a lot of this material, is Tractable
Structures in Constraint Sastifaction, in Handbook of
Constraint Processing,
P. Van Beek, F Rossi, and T Walsh (Eds). To appear 2006.
Introduction to
Reinforcement Learning
Andy Barto
April 18-20, 2006
9h-13h
Room 102
Slides: Lecture 1, Lecture 2, Lecture 3
Rina's part will be based mostly on her book Constraint
Processing, while Andy's on his book with Rich Sutton,
Introduction
to Reinforcement Learning. Within the time constraints, they
will also discuss more recent work.
All the lectures will be held in the building of the Department of
Technology,
Universitat Pompeu Fabra, next to the Estacion de Francia,
Barcelona.
Room 102.
Those who are planning to attend and are not enrolled in the UPF PhD
course (official name: Topics in AI); should send a note to the
coordinator.
Hector
Geffner
hector.geffner@upf.edu
11/05