Patent
1994-12-23
1997-07-01
Downs, Robert W.
395 25, G06F 1518
Patent
active
056446810
ABSTRACT:
A learning method for a neural network, in which at least a portion of the interconnection strength between neurons takes discrete values, includes the steps of updating an imaginary interconnection strength taking continuous values by using the quantity of update of the interconnection strength which has been calculated by using the discrete interconnection strength, and discretizing the imaginary interconnection strength so as to provide a novel interconnection strength.
REFERENCES:
patent: 4918618 (1990-04-01), Tomlinson, Jr.
patent: 4947482 (1990-08-01), Brown
patent: 4994982 (1991-02-01), Duranton et al.
patent: 5008833 (1991-04-01), Agranat et al.
patent: 5052043 (1991-09-01), Gaborski
Shuichi Tai, et al., "Recognition of 26 character-alphabet using a dynamic opto-electronic neural network", Int'l Joint Conference on Neural Networks, Jan. 15-19, 1990, Washington, D.C.
Nabil H. Farhat, "Optoelectronic analogs of self-programming neural nets: architecture and methodologies for implementing fast stochastic learning by simulated annealing", Applied Optics, vol. 26, No. 23, Dec. 1, 1987, pp. 5093-5103.
Hans P. Graf et al., "VLSI Implementation of a Neural Network Model", Computer, Mar. 1988, pp. 41-49.
Marcin Skubiszewski, "A Hardware Emulator For Binary Neural Works", International Neural Network Conference, Jul. 9-13, 1990.
Patrick A. Shoemaker et al., "Back Propagation Learning With Trinary Quantization of Weight Updates", Neural Networks, vol. 4, No. 2, 1991, pp. 231-241.
A de Callatay, "All-or-None Connections or Variable Weights?", Parallel Processing in Neural Systems and Computers, 1990, p. 233-236.
Donald R. Tveter, "Better Speed Through Integers", AI Expert Nov. 1990, pp. 40-46.
E. Fiesler, et al., "A weight discretization paradigm for optical neural networks", SPIE, vol. 1281 Optical Interconnections and Networks (1990), pp. 164-173.
Jack Raffel, et al., "A Generic Architecture for Wafer-Scale Neuromorphic Systems", M.I.T. Lincoln Laboratory, pp. 99-111.
Rumelhart, "Training Hidden Units: The Generalized Delta Rule", The Massachusetts Institute of Technology, 1988, pp. 121-160.
Alan F. Murray and Anthony V.W. Smith, "Asynchronous VLSI Neural Networks Using Pulse-Stream Arithmetic", Reprinted from IEEE Journal of Solid State Circuits, vol. 23, Jun. 1988.
Masaya Oita, et al., "Character recognition using a dynamic optoelectronic neural network with unipolar binary weights," Optics Letters, vol. 15, No. 21, Nov. 1, 1990, pp. 1227-1229.
Von Lehman, et al., "Influence of interconnection weight discretization and noise in an optoelectronic neural network", Optics Letters, vol. 14, No. 17, Sep. 1, 1989, pp. 928-930.
E. Fiesler, A. Choudry, & H.J. Caulfield, "Weight discretization in backward error propagation neural networks", presented at the First Annual Meeting of the International Neural Network Society, Boston, MA, Sep. 5-10, 1988, abstract: Neural networks, vol. 1, supplement 1, p. 380, 1988.
A Weight discretization paradigm for optical Neural Networks E Fiesler Mar. 13-15 1990, pp. 164-173 SPIE vol. 1281 Nerocomputing.
Hecht Nielsen/Addison Wesley (1989).
Neural Computing, Theory and Practice
Wasserman/Van Nostrand Reinhold (1989).
Kojima Keisuke
Kyuma Kazuo
Shinnishi Toshio
Tai Shuichi
Takahashi Masanobu
Downs Robert W.
Mitsubishi Denki & Kabushiki Kaisha
LandOfFree
Learning method for neural network having discrete interconnecti does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Learning method for neural network having discrete interconnecti, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Learning method for neural network having discrete interconnecti will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-604462