True wisdom comes to each of us when we realize how little we understand about life, ourselves, and the world around us - Socrates

Neural Networks

Teaching -> Neural Networks 

Duration: Approximately 30 lecture hours

Prerequisites: Knowledge in Calculus, Linear Algebra and Matlab

Course Contents:
  1. Introduction
  2. Neuron Models and Network Architectures
  3. Perceptron Learning Rule
  4. Supervised Hebbian Learning
  5. Widrow-Hoff Learning
  6. Back Propogation
  7. Associative Learning
  8. Competitive Networks

  • To introduce the main concepts, techniques and applications in the field of Neural Networks.

Learning outcomes:
  • Have an understanding of the concepts and techniques of neural networks through the study of the most important neural network models.
  • Have a knowledge of sufficient theoretical background to be able to reason about the behaviour of neural networks.
  • Be able to evaluate whether neural networks are appropriate to a particular application.
  • Be able to apply neural networks to particular applications, and to know what steps to take to improve performance.
  • Be able to implement Neural Network algorithms in Matlab.

Method of Assessments: Explained in the class

Attendance: Only those who satisfy the requirement of attendance (at least 80%) at lectures are allowed to sit the end of semester examination.

Recommended  Reading:
  • M. Hagan, H. Demuth and M. Beale, Neural Network Design, PWS Publishing Company, 1996.
  • H. Demuth and M. Beale, Neural Network Toolbox For Use With Matlab, User's Guide, Version 4, The Mathworks Inc., July 2002.