Paper Abstract and Keywords |
Presentation |
2023-08-08 15:15
Design and Implementation of Neuron Circuit Using Adiabatic Quantum-Flux-Parametron Logic Tomoharu Yamauchi, Hao San (Tokyo City Univ.), Naoki Takeuchi (AIST/Yokohama National Univ.), Nobuyuki Yoshikawa (Yokohama National Univ.), Olivia Chen (Tokyo City Univ.) SCE2023-11 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
Adiabatic quantum-flux-parametron (AQFP) logic is a promising technology for future energy-efficient,high performance information processing systems. It is a low power dissipation circuits in superconducting digital circuits.In this paper,we proposed a binary neural network (BNN) to achieve low power dissipation and small area, and designed and implemented a neuron circuit that introduces analog operations using magnetic coupling-based confluence. To effectively solve the memory problem in neural networks using superconducting integrated circuits. As an operational demonstration, we designed and implemented 4, 8 and 16 input neuron circuits using AIST $mathrm{10,kA/cm^2}$ 4-layer Nb process,and confirmed their operation from low-speed measurements at $mathrm{100,kHz}$. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
adiabatic logic / superconducting digital circuits / AQFP / binary neural network / / / / |
Reference Info. |
IEICE Tech. Rep., vol. 123, no. 153, SCE2023-11, pp. 53-57, Aug. 2023. |
Paper # |
SCE2023-11 |
Date of Issue |
2023-08-01 (SCE) |
ISSN |
Online edition: ISSN 2432-6380 |
Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
Download PDF |
SCE2023-11 |
|