Method for adjusting network parameters in a multi-layer percept

Details Method for adjusting network parameters in a multi-layer percept Method for adjusting network parameters in a multi-layer percept

1995-02-21

1997-11-18

Davis, George B.

395 24, G06F 1518

Patent

active

056896220

ABSTRACT:
A perceptron device has processing elements arranged in an input layer, one or more hidden layers, and an output layer. The training method of the perceptron device uses the back propagation method. In order to increase learning speed, the learning rate for the updates of a particular processing element is a function f(M,N,K) of the number N of inputs to that element, the number K of outputs of that element, and the number N of inputs to the next layer. Function f increases with increasing M, and decreases with increasing K and N.

REFERENCES:
patent: 4874963 (1989-10-01), Alspector
patent: 5003490 (1991-03-01), Castelaz et al.
patent: 5033006 (1991-07-01), Ishizuka et al.
patent: 5046020 (1991-09-01), Filkin
patent: 5056037 (1991-10-01), Eberhardt
patent: 5107454 (1992-04-01), Niki
patent: 5109351 (1992-04-01), Simar, Jr.
patent: 5129039 (1992-07-01), Hiraiwa
patent: 5195169 (1993-03-01), Kamiya et al.
"On the use of Backpropagation in Associative Reinforcement Learning", Williams et al, IEEE, Int. Conf. on Neural Networks Jul. 24-27, 1988.
"Experiments on Neural Net Recognition of Spoken and Written Text", Burr et al., IEEE Trans. on Acoustics, Speech and Signal Processing, 1988.
"Accelerating the Convergence of the Back-Propagation Method", Vogl et al, Biol. Cybern, 1988.
"Learning Algorithm for Perceptrons using Back-Propagation with Selective Updates", Huang et al, 1990 IEEE.
"A Simple Procedure for Pruning Back-Propagation Trained Neural Networks", Karnin et al, 1990 IEEE, Tran-on-n-n, Jun. 1990.
"Neural Networks, Part 2", Wasserman et al, 1988 IEEE.
"Increased Rates of Convergence Through Learning Rate Adapation" Jacobs Neural Networks, vol. 1, No. 4, 1988.
Castelaz et al,"Neural Networks in Defense Applications", IEEE Inter. Conf. on Neural Networks, 1988.

Affiliated with

Also associated with

Rating

Method for adjusting network parameters in a multi-layer percept does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Method for adjusting network parameters in a multi-layer percept, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for adjusting network parameters in a multi-layer percept will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-1572394