DocumentCode :
2636504
Title :
Structured backpropagation network
Author :
Cheng, L.M. ; Mak, H.L. ; Cheng, L.L.
Author_Institution :
Dept. of Electron. Eng., City Polytech. of Hong Kong, Kowloon, Hong Kong
fYear :
1991
fDate :
18-21 Nov 1991
Firstpage :
1641
Abstract :
A novel structured backpropagation network is described. It is different from the classic network as it requires prior knowledge of the problems that the network is dealing with by constructing an appropriate structure to match the problem. This approach can reduce training time and minimize the convergent problems that exist in the classic approaches. This new hybrid network architecture is superior to the traditional approach as this approach is extremely adaptable and expandable to allow the incorporation of mixed network types such as Hopfield, Boltzmann, and backpropagation. To illustrate how this structure network operates, an example in power control systems is shown
Keywords :
learning systems; neural nets; power system computer control; Boltzmann nets; Hopfield nets; learning systems; neural nets; power control systems; structured backpropagation network; Artificial neural networks; Backpropagation; Cities and towns; Control system analysis; Neurons; Pattern recognition; Power control; Power system modeling; Predictive models; Signal processing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 1991. 1991 IEEE International Joint Conference on
Print_ISBN :
0-7803-0227-3
Type :
conf
DOI :
10.1109/IJCNN.1991.170640
Filename :
170640
Link To Document :
بازگشت