1990-11-28
1995-08-01
MacDonald, Allen R.
395 23, G06F 1518
Patent
active
054386454
ABSTRACT:
A neural network includes a monitor circuit for monitoring an output value of each neuron and a weighting calculation circuit for calculating a weighting between neurons in accordance with designation from the monitor circuit in order to determine the weightings between the neurons in accordance with an energy function expressed in the form of a sum of a constrain condition and an evaluation function. The monitor circuit monitors the output values of the neurons and transmits, to the weighting calculation circuit, information representing an unsatisfactory state in which a distribution of the output values does not satisfy the constrain condition. The weighting calculation circuit changes the weightings between the neurons, e.g., doubles the weightings, every time it receives this information.
REFERENCES:
patent: 3103648 (1961-08-01), Hartmanis
patent: 3308441 (1963-07-01), Dusheck, Jr.
patent: 3348214 (1965-06-01), Barbetta
patent: 3950733 (1976-04-01), Cooper et al.
patent: 4651239 (1987-03-01), Andes et al.
patent: 4660166 (1987-04-01), Hopfield
patent: 4807168 (1989-02-01), Moopenn et al.
patent: 4858147 (1989-08-01), Conwell
patent: 4866645 (1989-09-01), Lish
patent: 4912655 (1990-05-01), Wood
patent: 4926064 (1990-05-01), Tapang
patent: 4943931 (1990-07-01), Allen
patent: 4947482 (1990-08-01), Brown
patent: 4951239 (1990-08-01), Andes et al.
patent: 5014219 (1991-05-01), White
patent: 5016211 (1991-05-01), Jeong
patent: 5023045 (1991-06-01), Watanabe et al.
patent: 5033006 (1991-07-01), Ishizuka et al.
patent: 5045713 (1991-09-01), Shima
patent: 5056037 (1991-10-01), Eberhardt
patent: 5107454 (1992-04-01), Niki
patent: 5195169 (1993-03-01), Kamiya et al.
"Neural Computing, Theory and Practice" Wasserman et al, Apr. 1989.
"Design of a Neural-Based A/D Convertor Using Modified Hopfield Network", Lee et al, IEEE Journal of Solid-State Circuits, Aug. 24, 1989.
"Back-Propagation Learning and Nonidealities in Analog Neural Network Hardware" Frye et al, Transaction on Neural Network, Jan. 1991.
"Neural Network, part 2" IEEE Expert, Wassermann et al, 1988.
"Neural Selective Processing and learning", Patrice Gelband and Edison Tse, IEEE Inter. Conf. on Neural Networks, Jul. 1988, pp. I-417-I-424.
"Computing with Neural Circuits: A Model", John J. Hopfield and David W. Tank, Aug. 1986.
IEEE Transactions on Information Theory vol. 35, No. 2, Mar. 1989, New York US pp. 263-280, "Classification Performance of Hopfield Neural Network Based on Hebbian-like Learning Rule".
The 15th Conference of Electrical & Electronics Engineers in Israel 7 Apr. 1987, Tel Aviv, Israel pp. 1-3, "A Training Algorithm for Neural Network".
Neural Networks from Models to Applications, E.S.P.C.I. 1988, Paris, France pp. 207-216, "Training Recurrent Networks".
Biological Cybernetics, vol. 52, pp. 141-152 "Neural Computation of Decisions in Optimization Problems": Hopfield and Tank; 1985.
Proceedings of the National Academy of Sciences 79:2544-2558 "Neural Networks and Physical Systems with Emergent Collective Computational Abilities", 1982 J. J. Hopfield, pp. 460-464.
Proceedings of the National Academy of Sciences 81:3088-3092 "Neurons with Graded Response have Collective Computational Properties like those of Two-State Neurons", 1984 J. J. Hopfield, pp. 579-583.
Davis George
Kabushiki Kaisha Toshiba
MacDonald Allen R.
LandOfFree
Neural network which uses a monitor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Neural network which uses a monitor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Neural network which uses a monitor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-739129