Module 12412 (2003)

Syllabus page 2003/2004

06-12412
Introduction to Neural Computation
Level 4/M

Peter Tino
10 credits in Semester 1

Links | Outline | Aims | Outcomes | Prerequisites | Teaching | Assessment | Books | Detailed Syllabus

The Module Description is a strict subset of this Syllabus Page. (The University module description has not yet been checked against the School's.)

Changes and updates

Small change made to final Learning Outcome (29 Sep 2003)

Relevant Links

See Introduction to Neural Computation Web-page for module material and further useful links.

Outline

Basic Neurobiology; Neural Networks; Single Neuron Models; Single Layer Perceptrons; Multi-Layer Perceptrons; Recurrent Networks; Radial Basis Function networks; Committee machines; Kohonen networks; Applications of neural networks.

Aims

The aims of this module are to:

  • introduce some of the fundamental techniques and principles of neural computation
  • investigate some common models and their applications

Learning Outcomes

On successful completion of this module, the student should be able to: Assessed by:
1understand the relation between real brains and simple artificial neural network models Examination
2describe and explain the most common architectures and learning algorithms for Multi-Layer Perceptrons, Recurrent Networks, Radial-Basis Function Networks, Committee Machines, and Kohonen Self-Organising Maps Examination
3explain the learning and generalisation aspects of neural computation Examination, assignment
4demonstrate an understanding of the implementational issues for common neural network systems Examination, assignment
5demonstrate an understanding of the practical considerations in applying neural computation to real classification and regression problems Examination, assignment

Restrictions, Prerequisites and Corequisites

Restrictions:

Excluded combination with 06-02360 (Introduction to Neural Networks).

Prerequisites:

None

Co-requisites:

None

Teaching

Teaching Methods:

2 hrs of lectures per week plus labs

Contact Hours:

24

Assessment

  • Supplementary (where allowed): As the sessional assessment
  • 2 hr open book examination (70%), continuous assessment (30%). Resit (where allowed) by examination only with the continuous assessment mark carried forward.

Recommended Books

TitleAuthor(s)Publisher, Date
Neural Networks: A Comprehensive FoundationS HaykinPrentice Hall, 1999
Neural Networks for Pattern RecognitionC M BishopOxford University Press, 1995
An Introduction to Neural NetworksK GurneyRoutledge, 1997
An Introduction to the Theory of Neural ComputationJ Hertz, A Krogh & R G PalmerAddison Wesley, 1991
Introduction to Neural NetworksR Beale & T JacksonIOP Publishing, 1990

Detailed Syllabus

  1. Introduction to Neural Networks and their History.
  2. Biological Neurons and Neural Networks, Artificial Neurons.
  3. Networks of Neurons, Single Layer Perceptrons (SLPs).
  4. Learning and Generalization in SLPs.
  5. Hebbian Learning, Gradient Descent Learning.
  6. The XOR Problem, Linear Separability, Multi-Layer Perceptrons (MLPs).
  7. The Back-Propagation Learning Algorithm and its Variations.
  8. Other optimization algorithms - Line Searches, Conjugate Gradient.
  9. Bias and Variance, Improving Generalization.
  10. Constructive Algorithms, Pruning Algorithms.
  11. Applications of Multi-Layer Perceptrons.
  12. Recurrent Networks.
  13. Radial-Basis Function Networks.
  14. Applications of Radial-Basis Function Networks.
  15. Committee Machines.
  16. Kohonen Self-Organising Maps (SOMs).
  17. Learning Vector Quantisation (LVQ).
  18. Overview of More Advanced Topics.
  19. Summary and Review.

Last updated: 29 Sep 2003

Source file: /internal/modules/COMSCI/2003/xml/12412.xml

Links | Outline | Aims | Outcomes | Prerequisites | Teaching | Assessment | Books | Detailed Syllabus