DocumentCode
1597360
Title
Neural network model based on weights and unit-offset control
Author
Lim, Chunhwan ; Park, Jongan
Author_Institution
Dept. of Electron. Eng., Chosun Univ., Kwangju, South Korea
Volume
2
fYear
1996
Firstpage
1385
Abstract
In this paper, we propose an improved algorithm for a neural network. It consists of multilayer neural networks with control part and address memory part. The control part controls weights between the input layer and the hidden layer, controls unit offsets of the hidden layer, and sends the output data to the address memory part. The address memory part memorizes the output pattern of the hidden layer, compared with the input pattern and sends the learning data to the output layer after learning. Simulation results show that it is easy to control weights and unit-offset. Its convergence speed is fast
Keywords
backpropagation; multilayer perceptrons; address memory part; convergence speed; hidden layer; improved algorithm; input layer; multilayer neural networks; neural network model; unit-offset control; weights; Backpropagation algorithms; Convergence; Electronic mail; Equations; Error correction; Learning systems; Multi-layer neural network; Neural networks; Neurons; Weight control;
fLanguage
English
Publisher
ieee
Conference_Titel
Signal Processing, 1996., 3rd International Conference on
Conference_Location
Beijing
Print_ISBN
0-7803-2912-0
Type
conf
DOI
10.1109/ICSIGP.1996.566572
Filename
566572
Link To Document