Neural network elements

Details Neural network elements Neural network elements

1995-02-07

1996-07-16

Moore, David K.

395 21, 395 27, G06E 100, G06F 1518, G06G 700

Patent

active

055375120

ABSTRACT:
An analog neural network element includes one or more EEPROMs as analog, reprogrammable synapses applying weighted inputs to positive and negative term outputs which are combined in a comparator. In one embodiment a pair of EEPROMs is used in each synaptic connection to separately drive the positive and negative term outputs. In another embodiment, a single EEPROM is used as a programmable current source to control the operation of a differential amplifier driving the positive and negative term outputs. In a still further embodiment, an MNOS memory transistor replaces the EEPROM or EEPROMs. These memory elements have limited retention or endurance which is used to simulate forgetfulness to emulate human brain function. Multiple elements are combinable on a single chip to form neural net building blocks which are then combinable to form massively parallel neural nets.

REFERENCES:
patent: 4153984 (1979-05-01), Hsia
patent: 4188670 (1980-02-01), Hsia
patent: 4268328 (1981-05-01), Hsia
patent: 4376987 (1983-03-01), Hsia
patent: 4398248 (1983-08-01), Hsia et al.
patent: 4454648 (1984-06-01), Hsi
patent: 4961002 (1990-10-01), Tam et al.
patent: 4969021 (1990-11-01), Thakoor et al.
patent: 5047655 (1991-09-01), Chambost et al.
patent: 5053638 (1991-10-01), Furutani et al.
patent: 5220202 (1993-06-01), Isono et al.
patent: 5222193 (1993-06-01), Brooks et al.
patent: 5247606 (1993-09-01), Tam
Yukun Hsia, "Dangling Bond Trapping Model for Si.sub.3 N_.sub.4 Gate Dielectric", Symposium on Interconnect, Gate, and Dielectric Materials, Fairchild Microprocessor Division, Oct. 17, 1994, University of Santa Clara.
Yukun Hsia and Eden Y. C. Mei, "The Impact Of Processing Conditions On EPROM Asperities And Device Deprogramming" Fairchild Microprocessor Division.
Yukun Hsia, Eden Mei and Kia L. Ngai, "MNOS Retention-Endurance Characteristics Enhancement Using Graded Nitride Dielectric", presented at the 1982 International Conference on Solid State Devices, Tokyo, Japan, 1982; Japanese Journal of Applied Physics, vol. 22, Supplement 22-1, pp. 89-93, 1983.
Yukun Hsia, "Silicon Machine-A Reconfigurable On-The-Wafer Interconnect Architecture", 1982 GOMAC Paper, McDonnell Douglas Corporation.
Kia L. Ngai and Yukun Hsia, "Empirical Study of the Metal-Nitride-Oxide-Semiconductor Device Characterisics Deduced From a Microscopic Model of Memory Traps", Applied Physics Letters, vol. 41, No. 2, pp. 150-161, Jul. 15, 1982.
Yukun Hsia, Richard W. Rodgers and Roman Fedorak, "A Reconfigurable Interconnect For In-Silicon Electronic Assembly", Reprint: Proceedings, 32nd Electronic Components Conference, San Diego, CA May 10-12, 1992, IEEE, pp. 7-16, 1982.
Yukun Hsia, Gareth C. C. Chang and F. Dennis Erwin, "Adaptive Wafer Scale Integration", presented at the 1979 International Conference on Solid State Devices, Tokyo, Japan, Aug, 27-29, 1979, Japanese Journal of Applied Physics, vol. 19 Supp. 19-1, pp. 193-202, 1980.
Yukun Hsia, "MNOS LSI Memory Device Data Retention Measurements and Projections", IEEE Transactions On Electron Devices, vol. Ed-24, No. 5, May 1977.
Yukun Hsia, "Cross-Gate MNOS Memory Device", IEEE Transactions on Electron Devices, vol. ED-25, No. 8, Aug. 1978.
Yukun Hsia, "Application Of The Quasi-Linearization Principle In Production Circuit Design", IEEE Transactions Of The Professional Technical Group On Circuit Theory, vol. CT-11, No. 2, Jun. 1964.
T. F. Tao and Yukun Hsia, "Dopant Dependency Of Fermi Level Location In Heavily Doped Silicon", Applied Physics Letters, vol. 13, No. 9, pp. 291-292, Nov. 1, 1968.
Yukun Hsia and Folk-Hwa Wang, "Switching Waveform Prediction of a Simple Transistor Inverter Circuit", IEEE Tranasactions On Electron Devices, vol. ED-12, No. 12, pp. 626-631, Dec. 1965.
Yukun Hsia and K. L. Ngai, "MNOS Traps and Tailored Trap Distribution Gate Dielectric MNOS", presented at the 1979 International Conference on Solid State Defvices, Tokyo, Japan, Japanese Journal of Applied Physics, vol. 19, Supplement 19-1, pp. 245-248.
Sin et al, "EEPROM as an analog storage device, with particular applications in neural networks"; IEEE Transactions on Electron devices, vol. 39, iss. 6, pp. 1410-1419, Jun. 1992.
Holler et al, "An electrically trainable artificial neural network (ETANN) with 10240 `floating gate` synapses"; IJCNN, pp. 191-196 vol. 2, 18-22, Jun. 1989.
Kramer et al, "EEPROM device as a reconfigurable analog element for neural networks"; International electron devices meeting 1989. Technical Digest, pp. 259-262, 3-6 Dec. 1989.
Card et al, "EEPROM synapses exhibiting pseudo-hebbian plasticity"; Electronic Letters, vol. 25 iss. 12, pp. 805-806, 8 Jun. 1989.
Yonezu et al, "Optoelectronic adaptive neuro-device"; Electronic Letters, vol. 28, iss. 8, pp. 715-717, 9 Apr. 1992.

Affiliated with

Also associated with

Rating

Neural network elements 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 elements, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Neural network elements will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-1791431