Adaptive Systems

Professor Bernard Manderick

To introduce pattern classification and neural networks together with their applications. The student has to be able to:

• understand the basic ideas behind these techniques
• implement these techniques
• apply these techniques to simple problems
• evaluate their performance

BA computer science or BA mathematics option computer science or equivalent. The following mathematical background is required: Basic calculus, probability theory and linear algebra.

• Introduction and Methodological Issues
• Bayesian Decision Theory
• Maximum Likelihood and Bayesian Estimation
• Nonparametric Techniques
• Linear Discriminant Functions
• Multilayer Neural Networks

Chapters 1–6 and 9 of Duda, Richard O. , Hart, Peter E. and David G. Stork, Pattern Classification, 2nd Edition, Wiley Interscience, 2000.

Slides Exercise Session 1

Gaussian Sample Matlab Function Rename to “***.m”!

Example report

Slides Exercise Session 2

Slides Exercise Session 3

Slides SVM

• Bishop, Christopher M. Pattern Recognition and Machine Learning, Springer, 2006.
• Alpaydin, Ethem, Introduction to Machine Learning, MIT Press 2004.

During the course, the students get 4 assignments that they can solve in groups of two. The final exam consists of giving an oral presentation of the solutions to these assignments. The assignments can be made in groups of 2 people.

• Group 1: Lode Hoste / Ben Maene
• Group 2: Maarten Deville / Kevin Van Vaerenbergh
• Group 3: Kristof Van Moffaert / Bart Van Campenhout
• Group 4: Tim Brys
• Group 5: Wajdi Al-Halabi
• Group 6: Simon Knuijver
• Group 7: Mark Van Lokeren/Gilles Vanderveken

For each assignment a report needs to be handed in discussing the results. The deadlines for the reports are:

• k-nearest neighbors: 26/10/2009 02/11/2009
• perceptron: 16/11/2009 23/11/2009
• neural network: 14/12/2009 21/12/2009
• support vector machines: week before examination