Title :
Convergence suppression and divergence facilitation: new approach to prune hidden layer and weights of feedforward neural networks
Author :
Yasui, Syozo ; Malinowski, Aleksander ; Zurada, Jacek M.
Author_Institution :
Neurosystems Lab., Kyushu Inst. of Technol., Fukuoka, Japan
fDate :
30 Apr-3 May 1995
Abstract :
A pruning algorithm is devised for multilayer multi-output feedforward perceptron networks. The algorithm efficiently reduces the total number of hidden units and the number of weights in the output layer. Test examples include network pruning for the IRIS classifier and for the bitmap digit classifier
Keywords :
backpropagation; convergence; feedforward neural nets; multilayer perceptrons; pattern classification; IRIS classifier; bitmap digit classifier; convergence suppression; divergence facilitation; feedforward neural networks; hidden layer; multilayer multi-output feedforward perceptron networks; pruning algorithm; weights; Convergence; Network topology; Neural networks; Neurons; Wiring;
Conference_Titel :
Circuits and Systems, 1995. ISCAS '95., 1995 IEEE International Symposium on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-2570-2
DOI :
10.1109/ISCAS.1995.521466
