Title :
Pruning product unit neural networks
Author :
Ismail, A. ; Engelbrecht, A.P.
Author_Institution :
Dept. of Comput. Sci., Univ. of Western Cape, South Africa
fDate :
6/24/1905 12:00:00 AM
Abstract :
Selection of the optimal architecture of a neural network is crucial to ensure good generalization by reducing the occurrence of overfitting. While much work has been done to develop pruning algorithms for networks that employ summation units, not much has been done on pruning of product unit neural networks. The paper develops and tests a pruning algorithm for product unit networks, and illustrates its performance on several function approximation tasks
Keywords :
function approximation; learning (artificial intelligence); mean square error methods; neural net architecture; statistical analysis; function approximation; generalization; optimal architecture; overfitting; product unit neural networks; pruning algorithm; Africa; Application software; Computer architecture; Computer networks; Computer science; Econometrics; Multi-layer neural network; Neural networks; Process planning; Testing;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1005479
