Optical implementation of winner-take-all models of neural networks

E. F. Kobzev; Mikhail A. Vorontsov

doi:10.1117/12.47500

1 April 1991 Optical implementation of winner-take-all models of neural networks

E. F. Kobzev, Mikhail A. Vorontsov

Proceedings Volume 1402, USSR-CSFR Joint Seminar on Nonlinear Optics in Control, Diagnostics, and Modeling of Biophysical Processes; (1991) https://doi.org/10.1117/12.47500
Event: USSR-CSFR Joint Seminar on Nonlinear Optics in Control, Diagnostics and Modeling, 1990, Koshice-Prague, Czechoslovakia

Abstract

The so called WTA ( " winner take all" ) model appears to be one of the most interesting models of artificial neural networks. '' In networks of this kind only one single neuron specified by maximum (or minimum) initial activity becomes active as a result of cooperative transient process. In this work the optical implementations of WTA neural network are presented. The example of WTA neural network demonstrates that due to the existence of strong NL competition the problem of global extremum search can actually be solved. It should be mentioned that this problem seems to be quite complicated for digital computers. It is worth adding that these neural networks have multiple applications.

Citation Download Citation

E. F. Kobzev and Mikhail A. Vorontsov "Optical implementation of winner-take-all models of neural networks", Proc. SPIE 1402, USSR-CSFR Joint Seminar on Nonlinear Optics in Control, Diagnostics, and Modeling of Biophysical Processes, (1 April 1991); https://doi.org/10.1117/12.47500

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available