Title :
Maximum certainty approach to feedforward neural networks
Author :
Roberts, S.J. ; Penny, W.
Author_Institution :
Dept. of Electr. & Electron. Eng., Imperial Coll. of Sci., Technol. & Med., London, UK
fDate :
2/13/1997 12:00:00 AM
Abstract :
A Bayesian-based methodology is presented which leads to a data analysis system based around a committee of radial-basis function (RBF) networks. The authors show that this approach enables estimation of the uncertainty associated with system outputs. Systems with differing numbers of internal degrees of freedom (weights) may hence be compared using training data only
Keywords :
Bayes methods; data analysis; feedforward neural nets; learning (artificial intelligence); Bayesian-based methodology; RBF networks; data analysis system; feedforward neural networks; maximum certainty approach; radial-basis function networks; system output uncertainty; training data; uncertainty estimation; weights;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:19970211
