Title :
Pruning strategies for the MTiling constructive learning algorithm
Author :
Parekh, Rajesh ; Tang, Ju ; Honavar, Vasant
Author_Institution :
Dept. of Comput. Sci., Iowa State Univ., Ames, IA, USA
Abstract :
We present a framework for incorporating pruning strategies in the MTiling constructive neural network learning algorithm. Pruning involves elimination of redundant elements (connection weights or neurons) from a network and is of considerable practical interest. We describe three elementary sensitivity based strategies for pruning neurons. Experimental results demonstrate a moderate to significant reduction in the network size without compromising the network´s generalization performance
Keywords :
generalisation (artificial intelligence); learning (artificial intelligence); neural nets; pattern classification; redundancy; sensitivity analysis; MTiling learning algorithm; connection weights; constructive neural network; generalization; network pruning; pattern classification; redundant elements; sensitivity; Algorithm design and analysis; Artificial intelligence; Artificial neural networks; Computer science; Feeds; Learning; Network topology; Neural networks; Neurons; Pattern classification;
Conference_Titel :
Neural Networks,1997., International Conference on
Conference_Location :
Houston, TX
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.614199
