dc.contributor.author |
Soloviev I. |
|
dc.contributor.author |
Schegolev A. |
|
dc.contributor.author |
Klenov N. |
|
dc.contributor.author |
Bakurskiy S. |
|
dc.contributor.author |
Kupriyanov M. |
|
dc.contributor.author |
Tereshonok M. |
|
dc.contributor.author |
Shadrin A. |
|
dc.contributor.author |
Stolyarov V. |
|
dc.contributor.author |
Golubov A. |
|
dc.date.accessioned |
2019-01-22T20:32:49Z |
|
dc.date.available |
2019-01-22T20:32:49Z |
|
dc.date.issued |
2018 |
|
dc.identifier.issn |
0021-8979 |
|
dc.identifier.uri |
https://dspace.kpfu.ru/xmlui/handle/net/147651 |
|
dc.description.abstract |
© 2018 Author(s). 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. |
|
dc.relation.ispartofseries |
Journal of Applied Physics |
|
dc.title |
Adiabatic superconducting artificial neural network: Basic cells |
|
dc.type |
Article |
|
dc.relation.ispartofseries-issue |
15 |
|
dc.relation.ispartofseries-volume |
124 |
|
dc.collection |
Публикации сотрудников КФУ |
|
dc.source.id |
SCOPUS00218979-2018-124-15-SID85054280182 |
|