Details of Grant 

EPSRC Reference:	GR/K78607/01
Title:	NEURAL COMPUTING AS A SOFTWARE TECHNOLOGY
Principal Investigator:	Partridge, Professor D
Other Investigators:
Researcher Co-Investigators:
Project Partners:	Pre Nexus Migration University of Exeter
Department:	Engineering Computer Science and Maths
Organisation:	University of Exeter
Scheme:	Standard Research (Pre-FEC)
Starts:	01 August 1996	Ends:	31 December 1999	Value (£):	222,402
EPSRC Research Topic Classifications:	Software Engineering
EPSRC Industrial Sector Classifications:	Retail
Related Grants:
Panel History:
Summary on Grant Application Form
The proposed project is a development from the neural computing safety-critical research at Exeter: it is to apply a draft methodology, developed for the well defined, abstract safety-critical problems, to 'data-defined' problems - i.e. problems that are primarily defined by a data set and are lacking in underlying theory, such as human face recognition. These problems are both very common and difficult to tackle with conventional computing technology which demands an up-front specification as perquisite. We propose to use an inductive programming technique (i.e. training neural networks) and to employ a multiversion approach (i.e. a set of alternative records) to combat the approximating nature of this technology. Success rests on the existence (and effective exploitation) of 'diversity' in the set of alternative networks. We propose to develop a framework for the engineering of the required diversity and systematic exploitation of it. Most real-world computing problems are actually data-defined, some have sufficient associated theory to enable an accurate abstract specification to be developed, others do not. In the former case we would hope to demonstrate an alternative technology, which may be cheaper and/or sometimes better. In the latter case we hope to offer a computing technology where none currently exists. The potential benefits are impossible to circumscribe.
Key Findings
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Potential use in non-academic contexts
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Impacts
Description This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk Summary Date Materialised
Sectors submitted by the Researcher
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Project URL:
Further Information:
Organisation Website:	http://www.ex.ac.uk