Superconducting implementation of neural networks using fluxon pulses

Y. Mizugaki; K. Nakajima; Y. Sawada; T. Yamashita

doi:10.1109/77.233508

Superconducting implementation of neural networks using fluxon pulses

Y. Mizugaki, K. Nakajima, Y. Sawada, T. Yamashita

Tohoku University

Research output: Contribution to journal › Article › peer-review

5 Citations (Scopus)

Abstract

We fabricated neural-based superconducting integrated-circuits by using Nb/AlO_x/Nb Josephson junctions, and demonstrated the operation of 2-bit neural-based A/D converter which is one of the circuits solving optimization problems. We used fluxon pulses as neural impulses and a Josephson junction as a threshold element. The conductance values of resistors by which Josephson transmission lines are connected represent fixed synaptic strengths. The preliminary experimental result suggests that variable critical currents of dc-SQUID may provide synapses with variable strength.

Original language	English
Pages (from-to)	2765-2768
Number of pages	4
Journal	IEEE Transactions on Applied Superconductivity
Volume	3
Issue number	1 pt 4
DOIs	https://doi.org/10.1109/77.233508
Publication status	Published - 1993 Mar

ASJC Scopus subject areas

Electrical and Electronic Engineering
Physics and Astronomy (miscellaneous)

Access to Document

10.1109/77.233508

Cite this

@article{26d2fdfc9a674e7f99dd25f2a1cdde74,

title = "Superconducting implementation of neural networks using fluxon pulses",

abstract = "We fabricated neural-based superconducting integrated-circuits by using Nb/AlOx/Nb Josephson junctions, and demonstrated the operation of 2-bit neural-based A/D converter which is one of the circuits solving optimization problems. We used fluxon pulses as neural impulses and a Josephson junction as a threshold element. The conductance values of resistors by which Josephson transmission lines are connected represent fixed synaptic strengths. The preliminary experimental result suggests that variable critical currents of dc-SQUID may provide synapses with variable strength.",

author = "Y. Mizugaki and K. Nakajima and Y. Sawada and T. Yamashita",

year = "1993",

month = mar,

doi = "10.1109/77.233508",

language = "English",

volume = "3",

pages = "2765--2768",

journal = "IEEE Transactions on Applied Superconductivity",

issn = "1051-8223",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "1 pt 4",

}

TY - JOUR

T1 - Superconducting implementation of neural networks using fluxon pulses

AU - Mizugaki, Y.

AU - Nakajima, K.

AU - Sawada, Y.

AU - Yamashita, T.

PY - 1993/3

Y1 - 1993/3

N2 - We fabricated neural-based superconducting integrated-circuits by using Nb/AlOx/Nb Josephson junctions, and demonstrated the operation of 2-bit neural-based A/D converter which is one of the circuits solving optimization problems. We used fluxon pulses as neural impulses and a Josephson junction as a threshold element. The conductance values of resistors by which Josephson transmission lines are connected represent fixed synaptic strengths. The preliminary experimental result suggests that variable critical currents of dc-SQUID may provide synapses with variable strength.

AB - We fabricated neural-based superconducting integrated-circuits by using Nb/AlOx/Nb Josephson junctions, and demonstrated the operation of 2-bit neural-based A/D converter which is one of the circuits solving optimization problems. We used fluxon pulses as neural impulses and a Josephson junction as a threshold element. The conductance values of resistors by which Josephson transmission lines are connected represent fixed synaptic strengths. The preliminary experimental result suggests that variable critical currents of dc-SQUID may provide synapses with variable strength.

UR - http://www.scopus.com/inward/record.url?scp=0027556359&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027556359&partnerID=8YFLogxK

U2 - 10.1109/77.233508

DO - 10.1109/77.233508

M3 - Article

AN - SCOPUS:0027556359

VL - 3

SP - 2765

EP - 2768

JO - IEEE Transactions on Applied Superconductivity

JF - IEEE Transactions on Applied Superconductivity

SN - 1051-8223

IS - 1 pt 4

ER -

Superconducting implementation of neural networks using fluxon pulses

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this