Accession Number : ADA189266

Title :   Experiments with the Back Propagation Algorithm: A Systematic Look at a Small Problem.

Descriptive Note : Memorandum rept.,

Corporate Author : ROYAL SIGNALS AND RADAR ESTABLISHMENT MALVERN (ENGLAND)

Personal Author(s) : Bedworth, M D ; Bridle, J S

PDF Url : ADA189266

Report Date : 29 Jun 1987

Pagination or Media Count : 15

Abstract : A multi-layer perception is a type of feed neural network for which there exists a learning algorithm based on error back-propagation. Experimental results are presented of the use of this error back-propagation algorithm on a carefully selected, simple problem. The effects of varying the structure of the network, the number of hidden units, the size of the training set and the initial weight values have been investigated. A number of ways of analysing solutions to such a simple problem are demonstrated. Explanations of the observed behaviour are offered which may provide insights applicable to a range of problems.

Descriptors :   *NEURAL NETS, *SIGNAL PROCESSING, *PATTERN RECOGNITION, ALGORITHMS, LAYERS, LEARNING, NETWORKS, WEIGHTING FUNCTIONS, ARTIFICIAL INTELLIGENCE, MARKOV PROCESSES, SPEECH RECOGNITION, VISION, WEIGHTING FUNCTIONS, AUTOMATION, AUDITORY PERCEPTION, VISUAL PERCEPTION, GREAT BRITAIN

Subject Categories : Cybernetics

Distribution Statement : APPROVED FOR PUBLIC RELEASE