Abstract for robinson_cnnss92

Presented at the Cambridge Neural Network Summer School

PRACTICAL NETWORK DESIGN AND IMPLEMENTATION

Tony Robinson

September 1992

The aim of this paper is to put Artificial Neural Network techniques in perspective, discussing the practical issues of network design and implementation. Firstly, a comparison is made with other pattern matching techniques with an overview of the capabilities and complexity of these techniques. Some basic ground in multi-layer perceptrons is then covered so that a link may be made between these and Gaussian classifiers. In the process, some of the practical difficulties associated with gradient descent based training are covered, along with the popular remedies. Finally, the paper is concluded with a summary of the advantages of recurrent networks and a review of the architectures available.


(ftp:) robinson_cnnss92.ps.Z (http:) robinson_cnnss92.ps.Z

If you have difficulty viewing files that end '.gz', which are gzip compressed, then you may be able to find tools to uncompress them at the gzip web site.

If you have difficulty viewing files that are in PostScript, (ending '.ps' or '.ps.gz'), then you may be able to find tools to view them at the gsview web site.

We have attempted to provide automatically generated PDF copies of documents for which only PostScript versions have previously been available. These are clearly marked in the database - due to the nature of the automatic conversion process, they are likely to be badly aliased when viewed at default resolution on screen by acroread.