Share Email Print
cover

Proceedings Paper

Four-quadrant optical matrix vector multiplication machine as a neural network processor
Author(s): Shai Abramson; D. Saad; Emanuel Marom; Naim Konforti
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Optical processors for neural networks are primarily fast matrix-vector multiplication machines that can potentially compete with serial computers owing to their parallelism and their ability to facilitate densely connected networks. However, in most proposed systems the multiplication supports only two quadrants and is thus unable to provide bipolar neuron outputs for increasing network capabilities and learning rate. We propose and demonstrate an opto-electronic four quadrant matrix-vector multiplier that can be used for feedforward neural networks recall and learning. Experimental results obtained with common commercial components demonstrate a novel, useful, and reliable approach for four quadrant matrix-vector multiplication in general and for feedforward neural network training and recall in particular.

Paper Details

Date Published: 13 August 1993
PDF: 9 pages
Proc. SPIE 1972, 8th Meeting on Optical Engineering in Israel: Optoelectronics and Applications in Industry and Medicine, (13 August 1993); doi: 10.1117/12.151090
Show Author Affiliations
Shai Abramson, Tel-Aviv Univ. (Israel)
D. Saad, Tel-Aviv Univ. (Israel)
Emanuel Marom, Tel-Aviv Univ. (Israel)
Naim Konforti, Tel-Aviv Univ. (Israel)


Published in SPIE Proceedings Vol. 1972:
8th Meeting on Optical Engineering in Israel: Optoelectronics and Applications in Industry and Medicine

© SPIE. Terms of Use
Back to Top