Title :
The research on learning algorithm of binary neural networks
Author :
Hua, Qiang ; Zheng, Qi-Lun
Author_Institution :
Huizhou Univ., Guangdong, China
fDate :
6/24/1905 12:00:00 AM
Abstract :
HGLA (Hamming-graph learning algorithm) is an unexhausted learning algorithm of binary neural networks. It affords the methods of using hidden neurons to express the most number of samples; the formula to calculate the weights and threshold; and the output neuron needs no training, so it is optimized
Keywords :
graph theory; learning (artificial intelligence); neural nets; HGLA; Hamming-graph learning algorithm; binary neural networks; hidden neuron; Algorithm design and analysis; Binary codes; Hamming distance; Input variables; Mercury (metals); Neural networks; Neurons; Optimization methods; Power line communications;
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.2002.1005530
