Research output

Conference Paper (in Formal Publication) ()

A walsh analysis of multilayer perceptron function

Citation
Swingler K (2014) A walsh analysis of multilayer perceptron function In: Proceedings of the International Conference on Neural Computation Theory and Applications (IJCCI 2014), Setubal, Portugal: Science and Technology Publications. NCTA 2014: 6th International Conference on Neural Computation Theory and Applications, 22.10.2014 - 24.10.2014, Rome, Italy, pp. 5-14.

Abstract
The multilayer perceptron (MLP) is a widely used neural network architecture, but it suffers from the fact that its knowledge representation is not readily interpreted. Hidden neurons take the role of feature detectors, but the popular learning algorithms (back propagation of error, for example) coupled with random starting weights mean that the function implemented by a trained MLP can be difficult to analyse. This paper proposes a method for understanding the structure of the function learned by MLPs that model functions of the class f : f1;1gn ! Rm. The approach characterises a given MLP using Walsh functions, which make the interactions among subsets of variables explicit. Demonstrations of this analysis used to monitor complexity during learning, understand function structure and measure the generalisation ability of trained networks are presented.

Keywords
Multilayer Perceptrons; Walsh Functions; Network Function Analysis

StatusPublished
AuthorsSwingler Kevin
Publication date2014
URLhttp://www.scitepress.org/…0004974800050014
PublisherScience and Technology Publications
Place of publicationSetubal, Portugal
ISBN 978-989-758-054-3
LanguageEnglish

Journal
Ncta 2014 - Proceedings of the International Conference on Neural Computation Theory and Applications (2014)

© University of Stirling FK9 4LA Scotland UK • Telephone +44 1786 473171 • Scottish Charity No SC011159
My Portal