We developd a method of implementing a multiple-valued Hopfield network on electronic circuits by using single-electron circuit technology. The single-electron circuit shows quantized behavior in its operation because of the discrete tunnel transport of electrons. It can therefore be successfully used for implementing neuron operation of the multiple-valued Hopfield network. The authors developed a single-electron neuron circuit that can produce the staircase transfer function required for the multiple-valued neuron. A method for constructing a multiple-valued Hopfield network by combining the neuron circuits was also developed. A sample network was designed that solves an example of the quadratic integer-programming problem. And a computer simulation demonstrated that the sample network can converge to its optimal state that represents the correct solution to the problem.
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
Takashi YAMADA, Yoshihito AMEMIYA, "A Multiple-Valued Hopfield Network Device Using Single-Electron Circuits" in IEICE TRANSACTIONS on Electronics,
vol. E82-C, no. 9, pp. 1615-1622, September 1999, doi: .
Abstract: We developd a method of implementing a multiple-valued Hopfield network on electronic circuits by using single-electron circuit technology. The single-electron circuit shows quantized behavior in its operation because of the discrete tunnel transport of electrons. It can therefore be successfully used for implementing neuron operation of the multiple-valued Hopfield network. The authors developed a single-electron neuron circuit that can produce the staircase transfer function required for the multiple-valued neuron. A method for constructing a multiple-valued Hopfield network by combining the neuron circuits was also developed. A sample network was designed that solves an example of the quadratic integer-programming problem. And a computer simulation demonstrated that the sample network can converge to its optimal state that represents the correct solution to the problem.
URL: https://global.ieice.org/en_transactions/electronics/10.1587/e82-c_9_1615/_p
Copy
@ARTICLE{e82-c_9_1615,
author={Takashi YAMADA, Yoshihito AMEMIYA, },
journal={IEICE TRANSACTIONS on Electronics},
title={A Multiple-Valued Hopfield Network Device Using Single-Electron Circuits},
year={1999},
volume={E82-C},
number={9},
pages={1615-1622},
abstract={We developd a method of implementing a multiple-valued Hopfield network on electronic circuits by using single-electron circuit technology. The single-electron circuit shows quantized behavior in its operation because of the discrete tunnel transport of electrons. It can therefore be successfully used for implementing neuron operation of the multiple-valued Hopfield network. The authors developed a single-electron neuron circuit that can produce the staircase transfer function required for the multiple-valued neuron. A method for constructing a multiple-valued Hopfield network by combining the neuron circuits was also developed. A sample network was designed that solves an example of the quadratic integer-programming problem. And a computer simulation demonstrated that the sample network can converge to its optimal state that represents the correct solution to the problem.},
keywords={},
doi={},
ISSN={},
month={September},}
Copy
TY - JOUR
TI - A Multiple-Valued Hopfield Network Device Using Single-Electron Circuits
T2 - IEICE TRANSACTIONS on Electronics
SP - 1615
EP - 1622
AU - Takashi YAMADA
AU - Yoshihito AMEMIYA
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Electronics
SN -
VL - E82-C
IS - 9
JA - IEICE TRANSACTIONS on Electronics
Y1 - September 1999
AB - We developd a method of implementing a multiple-valued Hopfield network on electronic circuits by using single-electron circuit technology. The single-electron circuit shows quantized behavior in its operation because of the discrete tunnel transport of electrons. It can therefore be successfully used for implementing neuron operation of the multiple-valued Hopfield network. The authors developed a single-electron neuron circuit that can produce the staircase transfer function required for the multiple-valued neuron. A method for constructing a multiple-valued Hopfield network by combining the neuron circuits was also developed. A sample network was designed that solves an example of the quadratic integer-programming problem. And a computer simulation demonstrated that the sample network can converge to its optimal state that represents the correct solution to the problem.
ER -