Neural Networks

Module code: G5015
Level 6
15 credits in spring semester
Teaching method: Seminar, Lecture
Assessment modes: Coursework

To take this module you must already be able to write software in one appropriate programming language such as Java, C, Python, or Matlab. Basic knowledge of formal computational skills is also a prerequisite.

In recent years neural computing has emerged as a practical technology with applications in many fields. The majority of these applications are concerned with problems in pattern recognition, and make use of feed-­forward network architectures such as the multi­layer perceptron and the radial basis function network.

It is widely acknowledged that the successful application of neural computing requires a principled approach, and this module will use the recent advances in neural computing to explore neural networks in-depth. By concentrating on the pattern-recognition aspects of neural networks, the module will cover many important topics such as:

  • spiking neural networks
  • multi­layer perception
  • radial basis function network
  • support vector machines
  • competitive learning
  • independent component analysis.

You will also learn to use neural networks in solving real world problems.


The course assumes an ability to write software in one appropriate programming language (e.g. Java, C, Python, Matlab). Basic knowledge of formal computational skills is also assumed.

Module learning outcomes

  • Demonstrate knowledge of the basic concepts of artificial neural networks.
  • Systematically understand how artificial neural networks (including fundamental techniques from dimensionality reduction) can be deployed as a general methodology for function approximation.
  • Have a detailed understanding of Multi-Layer Perceptrons (MLP) and Radial Basis Function (RBF) networks, and demonstrate knowledge of the fundamental principles of Support Vector Machines (SVM).
  • Be in a position to develop and apply neural networks to real-world problems, including pre-processing the data and incorporating prior knowledge wherever possible.