Modular feedforward neural network architecture with learning

Details Modular feedforward neural network architecture with learning Modular feedforward neural network architecture with learning

1995-04-28

1997-11-18

Downs, Robert W.

395 23, G06F 1518

Patent

active

056896212

ABSTRACT:
A first feedforward network receives an input vector signal and a plurality of weight signals and forms an output vector signal based thereupon. A second feedforward network, substantially identical to the first feedforward network, receives a first learning vector signal and the weight signals and forms a learning output vector based thereupon. A weight updating circuit generates the weight signals in accordance with a back propagation updating rule. The weight signals are updated based upon the first learning vector signal and a second learning vector signal in response to receiving a first predetermined layer signal, and are updated based upon the first learning vector signal, the second learning vector signal, and the learning output vector signal in response to receiving a second predetermined layer signal. A back propagation feedback network receives the second learning vector signal and the weight signals and generates a back propagated error vector signal based thereupon. The feedforward networks, the weight updating circuit, and the back propagation feedback network are implemented as analog circuits on a single integrated circuit.

REFERENCES:
patent: 5055897 (1991-10-01), Canepa et al.
patent: 5148514 (1992-09-01), Arima et al.
patent: 5214743 (1993-05-01), Asai et al.
patent: 5222193 (1993-06-01), Brooks et al.
patent: 5241509 (1993-08-01), Jousselin et al.
patent: 5274748 (1993-12-01), Boser et al.
patent: 5276773 (1994-01-01), Knauer et al.
patent: 5293457 (1994-03-01), Arima et al.
patent: 5293458 (1994-03-01), Chung et al.
patent: 5298796 (1994-03-01), Tawel
patent: 5317675 (1994-05-01), Ikehara
patent: 5319738 (1994-06-01), Shima et al.
patent: 5333241 (1994-07-01), Furuta et al.
patent: 5353382 (1994-10-01), Yariv et al.
36th Midwest Symposium On Circuit & Systems Proceedings, Aug. 16-18, 1993, Detroit,MI, "A Modular Analog Chip For Feedforward Networks With On-Chip Learning" Oh & Salam.
Proceedings Of 22nd Symposium On The Interface, May 16-19, 1990, "Learning Algorithms For Artificial Neural Nets For Analog Circuit Implementation", Fathi M.A. Salam.
"Analog CMOS Implementation Of Neural Network For Adaptive Signal Processing", Oh & Salam, May/Jun. 1994.
S. Garth, "A Chipset for High Speed Simulation of Neural Network Systems," 1987 IEEE Conf. on Neural Networks, pp. III-443 to III-452, Dec. 1987.
E. Fiesler et al., "A Weight Discretization Paragigm for Optical Neural Networks," SPIE vol. 1281 Optical Interconnections and Networks, pp. 164-173, Mar. 1990.

Affiliated with

Also associated with

Rating

Modular feedforward neural network architecture with learning does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Modular feedforward neural network architecture with learning, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Modular feedforward neural network architecture with learning will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-1572382