Knowledge in multivariate analysis, linear algebra, probability and differential equations is assumed.
Students are expected to acquire and develop the following habilities, taking into account this is a `scientific´ subject.
1. Basic general knowledge on the subject.
2. Capacity to analyse specific problems and synthesise acquired knowledge.
3. Specific problem solving.
4. Oral and written communication in English.
5. Capacity to apply knowledge on practical situation.
6. Capacity to learn and broaden knowledge.
7. Ability to work autonomously.
8. Research skills.
1. Introduction
2. Neural Network dynamics
3. Representation Capabilities of Neural Networks
4. Associative memories
5. Supervised learning
6. Unsupervised learning
7. Reinforcement learning
8. Boltzmann machines
9. Cellular Neural Networks.
The basic teaching method is through traditional classroom lessons where the theoretical fundamentals are developed and a few examples given.
From that, students must prepare an original research work using the techniques described in class. This research work should be written in congress paper form and due to public presentation.
Evaluation methods are:
J. Classroom participation.
M. Other: original research work presented in congress paper form.
The final mark is given according to participation in class and the quality of the research work, which shall be reviewed by the instructor as well as by classmates. The following items shall be taken into account:
- How original and rigorous the paper is.
- The quality of the writing and the public presentation.
- The quality and accuracy of the classmate review.
All course objectives are evaluated simultaneously by method M.
J.Hertz, A.Kroght, R.G.Palmer, Introduction to the theory of neural computation, Addison Wesley, 1991.
1. C.T.Lin, C.S.G.Lee, Neural Fuzzy Systems , Prentice-Hall, 1996.
2. M.H.Hassoun, Fundamentals of Artificial Neural Networks , a Bradford Book - MIT Press, 1995.
3. J.A.Freeman, D.M.Skapura, Redes Neuronales Addison-Wesley / Diaz de Santos, 1993.
4. B.Kosko, Neural Networks and Fuzzy Systems , Prentice-Hall, 1992.
5. A.Hyvärinen, E.Oja, Independent Component Analysis: Algorithms and applications, Neural Networks. 2000 May-Jun;13(4-5):411-430.