Doelstelling (Objectives)
At the end of this course, the student should be able to:- understand the principles of reasoning under uncertainty
- understand different numerical models for the representation of uncertainty, such as the CF model, the subjective Bayesian method, Bayesian belief networks, and possibly Dempster-Shafer theory
- have insight into model-based approaches to AI
- have insight into the pros and cons of learning models versus using expert knowledge
- have some experience in experimenting with computational intelligence systems to solve problems involving probability theory
|
Inhoud / Omschrijving (Contents)
Handling uncertain knowledge has been one of the central problems of AIresearch during the past 30 years. In the 1970s and 1980s uncertaintywas handled by means of formalisms that were linked to rule-basedrepresentation and reasoning methods. Since the 1990s probabilisticgraphical models, in particular Bayesian networks, are seen as theprimary formalisms to deal with uncertain knowledge. Both early andnew methods for represensenting uncertainty are studied in the course, where inparticular various aspects of Bayesian networks are covered.
|
Literatuur (Literature)
- P.J.F. Lucas and L.C. van der Gaag, Principles of Expert Systems, Addison-Wesley, Wokingham, 1991, Chapter 5.
- K.B. Korb and A.E. Nicholson, Bayesian Artificial Intelligence, Chapman & Hall, Boca Raton, 2004.
- R.G. Cowell, A.P. Dawid, S.L. Lauritzen and D.J.Spiegelhalter, Probabilistic Networks and Expert Systems,Springer, New York, 1999.
- F.V. Jensen, Bayesian Networks and Decision Graphs, Springer, New York, 2001.
|
Werkvormen (Teaching methods)
- 50 uur hoorcollege
- 10 uur individuele begeleiding
- 10 uur practicum
- 20 uur projectwerk
- 78 uur zelfstudie
|
Toetsingsvorm (Exam)
Written exam in addition to practical work.
|
Inschrijving ( Enrollment ) Through Student Portal |
Bijzonderheden (Extra information)
The course is part of the AI theme. - lectures
- tutorials
- practical
- seminar
|