This course will give an introduction to aspects of Neural Computation. This page will provide useful links to resources, and the various assignments and labs, as the course progresses.
Note that there will be a frequently asked questions slot for each lab, so that I can answer general questions and let everyone benefit from the typing...
|
Week
|
Frequently Asked Qustions
|
Topic
|
Student deliverables for following week
|
| 12 | Matlab - useful hints. | Introduction to Matlab | None |
| 13 |
Sample solution of basic neuron part |
Implementation of Basic perceptron. | Implement basic neural network in Matlab and visualise workings. |
| 14 | Lab 3 FAQ | Learning algorithms | Implement learning algorithm in Matlab and classify dataset. |
| 15 | Lab 4 FAQ | Back-propagation algorithm | ASSESSED HAND-IN worth 50% of practical marks |
| 16 | New dataset ideas. | ||
| 17 | Lab 6 FAQ | Transformation of inputs | ASSESSED - 10% Tickable |
| 18 | Unsupervised Learning | ASSESSED - 10% Tickable | |
| 19 | Applications | ASSESSED - 30% Hand-in Week 22 | |
| 20 |
A useful introductory paper on Neural networks is Jordan & Bishop.
Numerical Recipes in C book on-line. (Useful for background theory and example C programs. See Chapter 10: Minimization or Maximization of Functions for background on optimisation theory. Chapter 15 gives useful insight on modelling from data.)