Title :
A new fuzzy approach for setting the initial weights in a neural network
Author :
Lin, Yinghua ; Cunningham, George A., III
Author_Institution :
Dept. of Comput. Sci., New Mexico Tech, Socorro, NM, USA
Abstract :
We introduce “fuzzy curves” and use them to set the initial weights in a fuzzy-neural network. We show that our choice of initial weights and network structure yields networks that train very rapidly
Keywords :
fuzzy neural nets; learning (artificial intelligence); parallel architectures; architecture; fuzzy curves; fuzzy-neural network; initial weights; learning; network structure; Backpropagation; Computer science; Fuzzy neural networks; Fuzzy sets; Input variables; Intelligent networks; Joining processes; Marine vehicles; Neural networks; Neurons;
Conference_Titel :
Fuzzy Systems, 1994. IEEE World Congress on Computational Intelligence., Proceedings of the Third IEEE Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1896-X
DOI :
10.1109/FUZZY.1994.343692
