Optoelectronic implementation of multilayer neural networks in a single photorefractive crystal
(1990) In Optical Engineering 29(4). p.359-368- Abstract
We present a novel, versatile optoelectronic neural network architecture for implementing supervised learning algorithms in photorefractive materials. The system is based on spatial multiplexing rather than the more commonly used angular multiplexing of the interconnect gratings. This simple, single-crystal architecture implements a variety of multilayer supervised learning algorithms including mean field theory, back-propagation, and Marr-Albus-Kanerva style algorithms. Extensive simulations show how beam depletion, rescattering, absorption, and decay effects of the crystal are compensated for by suitably modified supervised learning algorithms.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/7b4a5783-0a65-40ff-a7e3-fd545cb39720
- author
- Peterson, Carsten LU ; Redfield, Stephen ; Keeler, James D. and Hartman, Eric
- organization
- publishing date
- 1990-04-01
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Optical Engineering
- volume
- 29
- issue
- 4
- pages
- 10 pages
- publisher
- SPIE
- external identifiers
-
- scopus:0025418857
- ISSN
- 0091-3286
- DOI
- 10.1117/12.55604
- language
- English
- LU publication?
- yes
- id
- 7b4a5783-0a65-40ff-a7e3-fd545cb39720
- date added to LUP
- 2019-05-14 16:07:32
- date last changed
- 2024-01-01 04:35:35
@article{7b4a5783-0a65-40ff-a7e3-fd545cb39720, abstract = {{<p>We present a novel, versatile optoelectronic neural network architecture for implementing supervised learning algorithms in photorefractive materials. The system is based on spatial multiplexing rather than the more commonly used angular multiplexing of the interconnect gratings. This simple, single-crystal architecture implements a variety of multilayer supervised learning algorithms including mean field theory, back-propagation, and Marr-Albus-Kanerva style algorithms. Extensive simulations show how beam depletion, rescattering, absorption, and decay effects of the crystal are compensated for by suitably modified supervised learning algorithms.</p>}}, author = {{Peterson, Carsten and Redfield, Stephen and Keeler, James D. and Hartman, Eric}}, issn = {{0091-3286}}, language = {{eng}}, month = {{04}}, number = {{4}}, pages = {{359--368}}, publisher = {{SPIE}}, series = {{Optical Engineering}}, title = {{Optoelectronic implementation of multilayer neural networks in a single photorefractive crystal}}, url = {{http://dx.doi.org/10.1117/12.55604}}, doi = {{10.1117/12.55604}}, volume = {{29}}, year = {{1990}}, }