We propose two kinds of architectures for implementing large-scale opto-electronic neural networks. These architectures are based on time- and frequency-division multiplexing (TDM and FDM) techniques, respectively, in which both the neuron state vector and the interconnection matrix are divided in the time- and frequency-domains. The computer simulations, which were carried out for the Hopfield associative memories in the neuron number of 400 and the memory number of 20, have shown their usefulness, providing almost the same recognition rate as the conventional architectures. Using the TDM technique, moreover, we experimentally demonstrated an opto-electronic implementation of the Hopfield associative memory. The experimental results showed that the number of the neurons was effectively increased. We further discuss how to construct the FDM system experimentally.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Jun OHTA, Masaya OITA, Shuichi TAI, Kunihiko HARA, Kazuo KYUMA, "Opto-Electronic Implementation of a Large-Scale Neural Network Using Multiplexing Techniques" in IEICE TRANSACTIONS on transactions,
vol. E73-E, no. 1, pp. 41-45, January 1990, doi: .
Abstract: We propose two kinds of architectures for implementing large-scale opto-electronic neural networks. These architectures are based on time- and frequency-division multiplexing (TDM and FDM) techniques, respectively, in which both the neuron state vector and the interconnection matrix are divided in the time- and frequency-domains. The computer simulations, which were carried out for the Hopfield associative memories in the neuron number of 400 and the memory number of 20, have shown their usefulness, providing almost the same recognition rate as the conventional architectures. Using the TDM technique, moreover, we experimentally demonstrated an opto-electronic implementation of the Hopfield associative memory. The experimental results showed that the number of the neurons was effectively increased. We further discuss how to construct the FDM system experimentally.
URL: https://global.ieice.org/en_transactions/transactions/10.1587/e73-e_1_41/_p
Copy
@ARTICLE{e73-e_1_41,
author={Jun OHTA, Masaya OITA, Shuichi TAI, Kunihiko HARA, Kazuo KYUMA, },
journal={IEICE TRANSACTIONS on transactions},
title={Opto-Electronic Implementation of a Large-Scale Neural Network Using Multiplexing Techniques},
year={1990},
volume={E73-E},
number={1},
pages={41-45},
abstract={We propose two kinds of architectures for implementing large-scale opto-electronic neural networks. These architectures are based on time- and frequency-division multiplexing (TDM and FDM) techniques, respectively, in which both the neuron state vector and the interconnection matrix are divided in the time- and frequency-domains. The computer simulations, which were carried out for the Hopfield associative memories in the neuron number of 400 and the memory number of 20, have shown their usefulness, providing almost the same recognition rate as the conventional architectures. Using the TDM technique, moreover, we experimentally demonstrated an opto-electronic implementation of the Hopfield associative memory. The experimental results showed that the number of the neurons was effectively increased. We further discuss how to construct the FDM system experimentally.},
keywords={},
doi={},
ISSN={},
month={January},}
Copy
TY - JOUR
TI - Opto-Electronic Implementation of a Large-Scale Neural Network Using Multiplexing Techniques
T2 - IEICE TRANSACTIONS on transactions
SP - 41
EP - 45
AU - Jun OHTA
AU - Masaya OITA
AU - Shuichi TAI
AU - Kunihiko HARA
AU - Kazuo KYUMA
PY - 1990
DO -
JO - IEICE TRANSACTIONS on transactions
SN -
VL - E73-E
IS - 1
JA - IEICE TRANSACTIONS on transactions
Y1 - January 1990
AB - We propose two kinds of architectures for implementing large-scale opto-electronic neural networks. These architectures are based on time- and frequency-division multiplexing (TDM and FDM) techniques, respectively, in which both the neuron state vector and the interconnection matrix are divided in the time- and frequency-domains. The computer simulations, which were carried out for the Hopfield associative memories in the neuron number of 400 and the memory number of 20, have shown their usefulness, providing almost the same recognition rate as the conventional architectures. Using the TDM technique, moreover, we experimentally demonstrated an opto-electronic implementation of the Hopfield associative memory. The experimental results showed that the number of the neurons was effectively increased. We further discuss how to construct the FDM system experimentally.
ER -