Electrical transmission or interconnection systems – Nonlinear reactor systems – Parametrons
Patent
1989-06-15
1992-09-08
Baker, Stephen M.
Electrical transmission or interconnection systems
Nonlinear reactor systems
Parametrons
307201, 364602, 364807, G06G 712, G06G 7184, G06J 100
Patent
active
051465426
ABSTRACT:
Neural nets using capacitive structures are adapted for construction in complementary metal-oxide-semiconductor integrated-circuit technology. Fully differential input amplifiers in each neural net layer apply their push-pull responses to synapse input signals via a pair of input lines. Each push-pull response is applied via a pair of capacitors of complementary capacitance values and an output line to the input of each of a plurality of non-linear output amplifiers in that neural net layer, which generate respective axon responses for that neural net layer. In certain of these neural nets, arrangements are made such that the capacitive structures are bilaterally responsive so that back-propagation calculations can be performed during training periods, to alter the relative values of capacitors in each pair thereof.
REFERENCES:
patent: 3890635 (1975-06-01), Engeler
patent: 3950733 (1976-04-01), Cooper et al.
patent: 3969636 (1976-07-01), Baertsch et al.
patent: 4156284 (1979-05-01), Engeler
patent: 4288858 (1981-09-01), Merola et al.
patent: 4903226 (1990-02-01), Tsividis
Soclof, S., Analog Integrated Circuits, pub. by Prentice-Hall, Inc., Englewood Cliffs, N.J. 07632, 1985, pp. 306, 333.
Hosticka, B. et al., "MOS Sampled Data Recursive Filters Using Switched Capacitor Integrators", IEEE Trans on Solid-State Circuits, vol. SC-12, No. 6, Dec. 1977, pp. 600-608.
Rumelhart, D. et al., Parallel Distributed Processing, vol. 1, copyright 1986 by The Massachusetts Inst. of Technology, pp. 45 -54.
Kub, F. et al., "Programmable Analog Synapses for Microelectronic Neural Networks Using a Hybrid Digital-Analog Approach", IEEE Int'l. Conf. on Neural Networks, 1988.
Bibyk, S., et al., "Issues in Analog VLSI and MOS Techniques for Neural Computing", in Analog VLSI Implementations of Neural Systems, ed. by C. Mead et al., copyright 1989 by Kluwer Academic Publishers, Proceedings of a workshop held May 8, 1989 in connection with the Int'l. Symp. on Circuits and Systems, pp. 103-133.
"An Introduction to Computing With Neural Nets", R. P. Lippmann, IEEE ASSP Magazine, April 1987, pp. 4-22.
"Switched-Capacitor Neural Networks", Y. P. Tsividis & D. Anastassiou, Electronics Letters, Aug. 27, 1987, vol. 23, No. 18, pp. 958-959.
"Analog MOS Integrated Circuits for Signal Processing", R. Gregorian & G. C. Temes, pp. 255-237, John Wiley & Sons, NYC, Chichester, Brisbane, Toronto, Singapore 1986.
Electronics letters, vol. 23, No. 24, Nov. 19, 1987, (Stevenage, Herts., GB) Y. Tsividis et al.: "Analogue Circuits for Variable-Synapse Electronic Neural Networks," pp. 1313-1314.
Electronics, vol. 45, No. 4, Feb. 14, 1972 R. G. Kostanty: "Doubling op amp Summing Power", pp. 73-75, see FIG. 4; p. 75, paragraph: AC Considerations.
Proceedings Neural Networks from Models to Applications, ESPCI, Paris, 1988, H. P. Graf et al.: "VLSI Neural Network for Fast Pattern Matrching", pp. 725-732.
International Journal of Electronics, vol. 64, No. 3, Mar. 1988, (London, GB) A. Chichocki et al.: "Switched-capacitor Function Generators", pp. 359-375.
Wescon 88, Conference Record, vol. 32, Nov. 1988, (North Hollywood, Calif., US), P. Hasler: "Implementing practical Neural Networks in Silicon", pp. SS/1/1-6.
Baker Stephen M.
Davis Jr. James C.
General Electric Company
Snyder Marvin
LandOfFree
Neural net using capacitive structures connecting output lines a 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 net using capacitive structures connecting output lines a, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Neural net using capacitive structures connecting output lines a will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-140525