Electrical transmission or interconnection systems – Nonlinear reactor systems – Parametrons
Patent
1988-02-11
1989-10-17
Hudspeth, David
Electrical transmission or interconnection systems
Nonlinear reactor systems
Parametrons
364807, 377 2, G06G 712
Patent
active
048749630
ABSTRACT:
A neuron network which achieves learning by means of a modified Boltzmann algorithm. The network may comprise interconnected input, hidden and output layers of neurons, the neurons being "on-off" or threshold electronic devices which are symmetrically connected by means of adjustable-weight synapse pairs. The synapses comprise the source-drain circuits of a plurality of paralleled FETs which differ in resistance or conductance in a binary sequence. The synapses are controlled by the output of an Up-Down counter, the reading of which is controlled by the results of a correlation of the states of the two neurons connected by the synapse pairs following the application of a set of plus and minus training signals to selected neurons of said network. A noise generator comprising a thermal noise source is provided for each neuron for the purpose of simulated annealing of the network.
REFERENCES:
patent: 3218475 (1965-11-01), Hiltz
patent: 3273125 (1966-09-01), Jakowatz
patent: 3287649 (1966-11-01), Rosenblatt
patent: 3374469 (1968-03-01), Connelly
patent: 3394351 (1968-07-01), Martin
patent: 3476954 (1969-11-01), Wennik et al.
patent: 3535693 (1970-10-01), Connelly
patent: 3950733 (1976-04-01), Cooper et al.
Widrow and Winter, "Neural Nets for Adaptive Filtering and Adaptive Pattern Recognition", IEEE Computer, Mar. 1988, pp. 25-39.
"A Neuromorphic VLSI Learning System", by Joshua Alspector and T. B. Allen, In: Advanced Research in VSLI: Proceedings of the 1987 Standford Conference, pp. 313-349.
"Absolute Stability of Global Pattern Formation and Parallel Memory Storage by Competitive Neural Networks", M. A. Cohen and S. Grossberg, IEEE, vol. SMC-13, No. 5, Sep./Oct. 1983, pp. 815-826.
"Neural Networks and Physical Systems with Emergent Collective Computational Abilities", J. J. Hopfield, Proc. Natl. Acad. Sci., vol. 79, Apr. 1962, pp. 2554-2558.
"NETtalk: A Parallel Network that Learns to Read Aloud", The Johns Hopkins University Electrical Engineering and Computer Science Technical Report, T. J. Sejnowski, C. R. Rosenberg, JHU/EECS-86/01.
"Contour Enhancement, Short Term Memory, and Constancies in Reverberating Neural Networks", Studies in Applied Mathematics, S. Grossberg, vol. LII, No. 3, Sep. 1973, pp. 213-357.
"A Logical Calculus of the Ideas Immanent in Nervous Activity", Bulletin of Mathematical Biophysics, W. S. McCulloch, W. Pitts, vol. 5, 1943, pp. 115-133.
"Distinctive Features, Categorical Perception, and Probability Learning: some Applications of a Neural Model", Psychological Review, J. A. Anderson et al., 1977, pp. 288-324.
"Learning Internal Representations by Error Propagation", Parallel Distributed Processing: Explorations in the Microstructures of Cognition, D. E. Rumelhart, et al., vol. 1, 1986, pp. 675-699.
"VLSI Implementation of a Neural Network Memory with Several Hundreds of Neurons", Proceedings of the Conference on Neural Networks for Computing American Institute of Physics, H. P. Graf et al., 1986, pp. 182-187.
"Skeleton Filters in the Brain", Parallel Models of Associate Memory, T. J. Sejnowski, 1981, pp. 189-212.
"A Learning Algorithm for Boltzmann Machines", Cognitive Science 9, D. H. Ackley et al., 1985, pp. 147-169.
"A Novel Associative Memory Implemented Using Collective Computation", Proceedings of the 1985 Chapel Hill Conference on Very Large Scale Integration, M. Sivilotti et al., pp. 329-342.
"Adaptive Switching Circuits", Institute of Radio Engineers, Western Electric Show and Convention, Convention Record, Part 4, B. Widrow, M. E. Hoff, 1960, pp. 96-104.
"Infinite-Ranged Methods of Spin-Glasses", Physical Review, S. Kirkpatrick et al., vol. 17, No. 11, Jun. 1, 1978, pp. 4384-4403.
Bell Communications Research Inc.
Falk James W.
Hudspeth David
LandOfFree
Neuromorphic learning networks does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Neuromorphic learning networks, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Neuromorphic learning networks will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1745143