Title :
An empirical comparison of node pruning methods for layered feedforward neural networks
Author :
Castellano, Giovanna ; Fanelli, Anna Maria ; Pelillo, Marcello
Author_Institution :
Dipartimento di Inf., Bari Univ., Italy
Abstract :
One popular approach to reduce the size of an artificial neural network is to prune off the hidden units after learning has taken place. This paper compares three different node pruning algorithms in terms of size and performance of the reduced network. Experimental results are reported and some useful conclusions are drawn.
Keywords :
feedforward neural nets; learning (artificial intelligence); performance evaluation; redundancy; hidden units; layered feedforward neural networks; learning; node pruning methods; redundant units; Artificial neural networks; Feedforward neural networks; Feedforward systems; Neural networks; Petroleum; Scattering; Testing;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.713922
