CS2NC19-Neurocomputation

Module Provider: School of Mathematical, Physical and Computational Sciences
Number of credits: 10 [5 ECTS credits]
Level:5
Terms in which taught: Autumn term module
Pre-requisites: CS1PR16 Programming and CS1AC16 Applications of Computer Science or PY1SN Introduction to Systems Neuroscience
Non-modular pre-requisites:
Co-requisites:
Modules excluded: CS2NN16 Neural Networks
Current from: 2019/0

Module Convenor: Prof Richard Mitchell

Email: r.j.mitchell@reading.ac.uk

Type of module:

Summary module description:

This module covers the theory and implementation of a few types of artificial neural network. In

addition, one network is used as a case study for object-oriented programming. Students are

expected to implement a neural network and apply it to real world problems.

Aims:

The module aims to describe in detail a mode of computation inspired by such biological

functionality, namely artificial neural networks. The module also demonstrates how such a network

can be programmed using object orientation.

This module also encourages students to develop a set of professional skills, such as programming and research where they find a data set and then apply it to their neural net and write up as a conference paper.

Assessable learning outcomes:

By the end of the module the student should be able to apply various neural network techniques to

'real-world' problems; and to program a simple neural network using the object oriented paradigm.

Additional outcomes:

Outline content:

Various neural network techniques are described, for some their implementation is provided, and

suitable applications discussed. Networks and techniques examined include data processing;

Single and Multi- Layer Perceptrons and associated learning methods; Radial Basis Function

networks, Weightless Neural Networks; Genetic Algorithms; Stochastic Diffusion Search and

Kohonen networks.

 

Associated with the lectures is an assignment whereby students use the object-oriented paradigm

to design and implement a neural network and then apply that network to a suitable problem.

Brief description of teaching and learning methods:

The module comprises 1 lecture per week, three lab practicals and an associated assignment.

Contact hours:

	Autumn	Spring	Summer
Lectures	10
Practicals classes and workshops	9
Guided independent study:	81
Total hours by term	100
Total hours for module	100

Summative Assessment Methods:

Method	Percentage
Set exercise	100

Summative assessment- Examinations:

Summative assessment- Coursework and in-class tests:

Three labs are used in which students implement a Neural Network

Feedback is provided after each lab to help students ensure their network works

Students then apply their network do data of their choice and write up as a research paper

Formative assessment methods:

Penalties for late submission:
The Module Convener will apply the following penalties for work submitted late:

where the piece of work is submitted after the original deadline (or any formally agreed extension to the deadline): 10% of the total marks available for that piece of work will be deducted from the mark for each working day[1] (or part thereof) following the deadline up to a total of five working days;

where the piece of work is submitted more than five working days after the original deadline (or any formally agreed extension to the deadline): a mark of zero will be recorded.

The University policy statement on penalties for late submission can be found at: http://www.reading.ac.uk/web/FILES/qualitysupport/penaltiesforlatesubmission.pdf
You are strongly advised to ensure that coursework is submitted by the relevant deadline. You should note that it is advisable to submit work in an unfinished state rather than to fail to submit any work.

Assessment requirements for a pass:

A mark of 40% overall.

Reassessment arrangements:

Examination only.

One 2-hour examination paper in August/September.

Additional Costs (specified where applicable):

Last updated: 8 April 2019