Graphical Models, Course at the Department of Computer ScienceUniversity of Helsinki

Graphical Models

''Graphical Models'' is a course on graphical probabilistic models. Graphical models are used for the structural aspects of a problem (including dependency, cause and relevance) and probabilistic techniques will be developed for their analysis and interpretation. Examples will be drawn from learning, information retrieval, location sensing, as well as several other areas.  Both theory and applications are covered from a perspective that combines both the statistical and the computer science aspects of the area.

Instructors: Sr. Research Scientist Wray Buntine, and Professor Petri Myllymäki
Complex Systems Computation Group
Course assistant: M.Sc. Teemu Roos

In addition to regular lectures the course involves project work. These pages will be updated during the course and the current schedule and topics are only tentative. Lectures take place at Lecture hall A320 at 10:15AM-1:00PM.

Course description
Registration
Project
Term paper
Material
Results (Intranet)

Course Schedule:

Fri Oct. 31 Lecture 1: Introduction: Graphs and Probability
Fri Nov. 7 Lecture 2: Independence Modelling
Fri Nov. 14 Lecture 3: Exact Inference
Fri Nov. 21 Lecture 4: Introduction to Modelling and Learning
Mon Nov. 24 Decide project topic (e-mail Teemu).
Fri Nov. 28 Lecture 5: Learning with Graphs
Fri Dec. 5 Lecture 6: Selected Applications
Wed Dec. 31 Project and Term paper deadline.
Mon Jan. 19 Grades.
Jan. Individual feedback sessions.


 

 Graphical Models
Last updated Jan. 16, 2004