ML Anthology

An Optoelectronic Architecture for Multilayer Learning in a Single Photorefractive Crystal

Carsten Peterson, Stephen Redfield, James D. Keeler, Eric Hartman
NeCo 1990 pp. 25-34
doi:10.1162/NECO.1990.2.1.25 /neco/1990/peterson1990neco-optoelectronic/

Abstract

We propose a simple architecture for implementing supervised neural network models optically with photorefractive technology. The architecture is very versatile: a wide range of supervised learning algorithms can be implemented including mean-field-theory, backpropagation, and Kanerva-style networks. Our architecture is based on a single crystal with spatial multiplexing rather than the more commonly used angular multiplexing. It handles hidden units and places no restrictions on connectivity. Associated with spatial multiplexing are certain physical phenomena, rescattering and beam depletion, which tend to degrade the matrix multiplications. Detailed simulations including beam absorption and grating decay show that the supervised learning algorithms (slightly modified) compensate for these degradations.

NeCo Semantic Scholar

Cite

Text

Peterson et al. "An Optoelectronic Architecture for Multilayer Learning in a Single Photorefractive Crystal." Neural Computation, 1990. doi:10.1162/NECO.1990.2.1.25

Markdown

[Peterson et al. "An Optoelectronic Architecture for Multilayer Learning in a Single Photorefractive Crystal." Neural Computation, 1990.](https://mlanthology.org/neco/1990/peterson1990neco-optoelectronic/) doi:10.1162/NECO.1990.2.1.25

BibTeX

@article{peterson1990neco-optoelectronic,
  title     = {{An Optoelectronic Architecture for Multilayer Learning in a Single Photorefractive Crystal}},
  author    = {Peterson, Carsten and Redfield, Stephen and Keeler, James D. and Hartman, Eric},
  journal   = {Neural Computation},
  year      = {1990},
  pages     = {25-34},
  doi       = {10.1162/NECO.1990.2.1.25},
  volume    = {2},
  url       = {https://mlanthology.org/neco/1990/peterson1990neco-optoelectronic/}
}