Applications of Neural Networks

Objectives

To consolidate knowledge about Problem Solving with ANNs based in Practical Cases of Engineering Problems of Moderate difficulty

General characterization

Code

3606

Credits

6.0

Responsible teacher

José Barahona da Fonseca

Hours

Weekly - 4

Total - 30

Teaching language

Português

Prerequisites

Having some experience with Matlab and the neural networks Toolbox and knowing the Hodgkin and Huxley model, the Rosenblatt perceptron and the Backpropagation algorithm.

Bibliography

1. M. T. Hagan, H. B. Demuth and  M. Beale,
 Neural Network Design,
1996, PWS PUBLISHING COMPANY, Boston, MA.

Teaching method

Theoretical-Pratical Classes with computational simulations using a datashow. It is assumed that each students brings his computer.

Evaluation method

5 Technical Reports that cover all the theory of the disciple, being the last an application of neural networks to a problem well known by the students.

Subject matter

1. Applications of Feedforward Neural Networks, 2. Applications of Recorrent Neural Networks, 3. Applications of Self Organizing Maps,  4. Applications of Support Vector Machines

Programs

Programs where the course is taught: