Title :
Sparselized higher-order neural network and its pruning algorithm
Author :
Wangchao, Li ; Yongbin, Wang ; Wenjing, Li ; Jie, Zhang ; Li, Jinyan
Author_Institution :
Hebei Univ. of Technol., Tianjin, China
Abstract :
In this paper, the fully-connected higher-order neuron and sparselized higher-order neuron are introduced, the mapping capabilities of the fully-connected higher-order neural networks are investigated, and that arbitrary Boolean function defined from (0,1)N can be realized by fully-connected higher-order neural networks is proved. Based on this, in order to simplify the networks´ architecture, a pruning algorithm of eliminating the redundant connection weights is also proposed, which can be applied to the implementation of sparselized higher-order neural classifier and other networks. The simulated results show the effectiveness of the algorithm
Keywords :
Boolean functions; function approximation; learning (artificial intelligence); neural nets; pattern classification; redundancy; Boolean function; function approximation; higher-order neural network; learning; mapping; neural classifier; pattern classification; pruning algorithm; redundant connection weights; Approximation methods; Boolean functions; Convergence; Inference algorithms; Neural networks; Neurons;
Conference_Titel :
Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on
Conference_Location :
Anchorage, AK
Print_ISBN :
0-7803-4859-1
DOI :
10.1109/IJCNN.1998.682292
