Adiabatic superconducting artificial neural network: Basic cells

Igor I. Soloviev (Corresponding Author), Andrey E. Schegolev, Nikolay V. Klenov, Sergey V. Bakurskiy, Mikhail Yu Kupriyanov, Maxim V. Tereshonok, Anton V. Shadrin, Vasily S. Stolyarov, Alexander A. Golubov

Research output: Contribution to journalArticleAcademicpeer-review

4 Citations (Scopus)

Abstract

We consider adiabatic superconducting cells operating as an artificial neuron and synapse of a multilayer perceptron (MLP). Their compact circuits contain just one and two Josephson junctions, respectively. While the signal is represented as magnetic flux, the proposed cells are inherently nonlinear and close-to-linear magnetic flux transformers. The neuron is capable of providing the one-shot calculation of sigmoid and hyperbolic tangent activation functions most commonly used in MLP. The synapse features both positive and negative signal transfer coefficients in the range ∼ (- 0.5, 0.5). We briefly discuss implementation issues and further steps toward the multilayer adiabatic superconducting artificial neural network, which promises to be a compact and the most energy-efficient implementation of MLP.

Original languageEnglish
Article number152113
JournalJournal of applied physics
Volume124
Issue number15
DOIs
Publication statusPublished - 21 Oct 2018

Fingerprint

self organizing systems
synapses
neurons
magnetic flux
cells
tangents
transformers
Josephson junctions
shot
activation
coefficients
energy

Cite this

Soloviev, I. I., Schegolev, A. E., Klenov, N. V., Bakurskiy, S. V., Kupriyanov, M. Y., Tereshonok, M. V., ... Golubov, A. A. (2018). Adiabatic superconducting artificial neural network: Basic cells. Journal of applied physics, 124(15), [152113]. https://doi.org/10.1063/1.5042147
Soloviev, Igor I. ; Schegolev, Andrey E. ; Klenov, Nikolay V. ; Bakurskiy, Sergey V. ; Kupriyanov, Mikhail Yu ; Tereshonok, Maxim V. ; Shadrin, Anton V. ; Stolyarov, Vasily S. ; Golubov, Alexander A. / Adiabatic superconducting artificial neural network : Basic cells. In: Journal of applied physics. 2018 ; Vol. 124, No. 15.
@article{89da74e8dd9741b4a969cf0205c1a298,
title = "Adiabatic superconducting artificial neural network: Basic cells",
abstract = "We consider adiabatic superconducting cells operating as an artificial neuron and synapse of a multilayer perceptron (MLP). Their compact circuits contain just one and two Josephson junctions, respectively. While the signal is represented as magnetic flux, the proposed cells are inherently nonlinear and close-to-linear magnetic flux transformers. The neuron is capable of providing the one-shot calculation of sigmoid and hyperbolic tangent activation functions most commonly used in MLP. The synapse features both positive and negative signal transfer coefficients in the range ∼ (- 0.5, 0.5). We briefly discuss implementation issues and further steps toward the multilayer adiabatic superconducting artificial neural network, which promises to be a compact and the most energy-efficient implementation of MLP.",
author = "Soloviev, {Igor I.} and Schegolev, {Andrey E.} and Klenov, {Nikolay V.} and Bakurskiy, {Sergey V.} and Kupriyanov, {Mikhail Yu} and Tereshonok, {Maxim V.} and Shadrin, {Anton V.} and Stolyarov, {Vasily S.} and Golubov, {Alexander A.}",
year = "2018",
month = "10",
day = "21",
doi = "10.1063/1.5042147",
language = "English",
volume = "124",
journal = "Journal of applied physics",
issn = "0021-8979",
publisher = "American Institute of Physics",
number = "15",

}

Soloviev, II, Schegolev, AE, Klenov, NV, Bakurskiy, SV, Kupriyanov, MY, Tereshonok, MV, Shadrin, AV, Stolyarov, VS & Golubov, AA 2018, 'Adiabatic superconducting artificial neural network: Basic cells' Journal of applied physics, vol. 124, no. 15, 152113. https://doi.org/10.1063/1.5042147

Adiabatic superconducting artificial neural network : Basic cells. / Soloviev, Igor I. (Corresponding Author); Schegolev, Andrey E.; Klenov, Nikolay V.; Bakurskiy, Sergey V.; Kupriyanov, Mikhail Yu; Tereshonok, Maxim V.; Shadrin, Anton V.; Stolyarov, Vasily S.; Golubov, Alexander A.

In: Journal of applied physics, Vol. 124, No. 15, 152113, 21.10.2018.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Adiabatic superconducting artificial neural network

T2 - Basic cells

AU - Soloviev, Igor I.

AU - Schegolev, Andrey E.

AU - Klenov, Nikolay V.

AU - Bakurskiy, Sergey V.

AU - Kupriyanov, Mikhail Yu

AU - Tereshonok, Maxim V.

AU - Shadrin, Anton V.

AU - Stolyarov, Vasily S.

AU - Golubov, Alexander A.

PY - 2018/10/21

Y1 - 2018/10/21

N2 - We consider adiabatic superconducting cells operating as an artificial neuron and synapse of a multilayer perceptron (MLP). Their compact circuits contain just one and two Josephson junctions, respectively. While the signal is represented as magnetic flux, the proposed cells are inherently nonlinear and close-to-linear magnetic flux transformers. The neuron is capable of providing the one-shot calculation of sigmoid and hyperbolic tangent activation functions most commonly used in MLP. The synapse features both positive and negative signal transfer coefficients in the range ∼ (- 0.5, 0.5). We briefly discuss implementation issues and further steps toward the multilayer adiabatic superconducting artificial neural network, which promises to be a compact and the most energy-efficient implementation of MLP.

AB - We consider adiabatic superconducting cells operating as an artificial neuron and synapse of a multilayer perceptron (MLP). Their compact circuits contain just one and two Josephson junctions, respectively. While the signal is represented as magnetic flux, the proposed cells are inherently nonlinear and close-to-linear magnetic flux transformers. The neuron is capable of providing the one-shot calculation of sigmoid and hyperbolic tangent activation functions most commonly used in MLP. The synapse features both positive and negative signal transfer coefficients in the range ∼ (- 0.5, 0.5). We briefly discuss implementation issues and further steps toward the multilayer adiabatic superconducting artificial neural network, which promises to be a compact and the most energy-efficient implementation of MLP.

UR - http://www.scopus.com/inward/record.url?scp=85054280182&partnerID=8YFLogxK

U2 - 10.1063/1.5042147

DO - 10.1063/1.5042147

M3 - Article

VL - 124

JO - Journal of applied physics

JF - Journal of applied physics

SN - 0021-8979

IS - 15

M1 - 152113

ER -

Soloviev II, Schegolev AE, Klenov NV, Bakurskiy SV, Kupriyanov MY, Tereshonok MV et al. Adiabatic superconducting artificial neural network: Basic cells. Journal of applied physics. 2018 Oct 21;124(15). 152113. https://doi.org/10.1063/1.5042147