Paper Abstract and Keywords |
Presentation |
2021-03-03 10:00
Energy Efficient Approximate Storing to MRAM for Deep Neural Network Tasks in Edge Computing Yoshinori Ono, Kimiyoshi Usami (SIT) VLD2020-67 HWS2020-42 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
On-chip learning is gaining attention in edge devices. In addition, a magnetic RAM (MRAM) is a promising memory technology for edge devices because of low leakage energy. However, the high write energy is a disadvantage of MRAM. For minimizing the write energy, we propose an approximate storing approach to MRAM for learning tasks of deep neural networks (DNN). The proposed approach writes the weight and bias data to MRAM approximately on each epoch with the fine-grained adjusted write time. Simulation results with image recognition DNN applications have demonstrated that the write energy can be reduced in the range from 9% to 37% with negligible (< 0.5%) accuracy loss. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
MRAM / Energy Minimization / Approximate Computing / Deep Learning / On-chip Learning / / / |
Reference Info. |
IEICE Tech. Rep., vol. 120, no. 400, VLD2020-67, pp. 1-6, March 2021. |
Paper # |
VLD2020-67 |
Date of Issue |
2021-02-24 (VLD, HWS) |
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 |
VLD2020-67 HWS2020-42 |
Conference Information |
Committee |
HWS VLD |
Conference Date |
2021-03-03 - 2021-03-04 |
Place (in Japanese) |
(See Japanese page) |
Place (in English) |
Online |
Topics (in Japanese) |
(See Japanese page) |
Topics (in English) |
Design Technology for System-on-Silicon, Hardware Security, etc. |
Paper Information |
Registration To |
VLD |
Conference Code |
2021-03-HWS-VLD |
Language |
Japanese |
Title (in Japanese) |
(See Japanese page) |
Sub Title (in Japanese) |
(See Japanese page) |
Title (in English) |
Energy Efficient Approximate Storing to MRAM for Deep Neural Network Tasks in Edge Computing |
Sub Title (in English) |
|
Keyword(1) |
MRAM |
Keyword(2) |
Energy Minimization |
Keyword(3) |
Approximate Computing |
Keyword(4) |
Deep Learning |
Keyword(5) |
On-chip Learning |
Keyword(6) |
|
Keyword(7) |
|
Keyword(8) |
|
1st Author's Name |
Yoshinori Ono |
1st Author's Affiliation |
Shibaura Institute of Technology (SIT) |
2nd Author's Name |
Kimiyoshi Usami |
2nd Author's Affiliation |
Shibaura Institute of Technology (SIT) |
3rd Author's Name |
|
3rd Author's Affiliation |
() |
4th Author's Name |
|
4th Author's Affiliation |
() |
5th Author's Name |
|
5th Author's Affiliation |
() |
6th Author's Name |
|
6th Author's Affiliation |
() |
7th Author's Name |
|
7th Author's Affiliation |
() |
8th Author's Name |
|
8th Author's Affiliation |
() |
9th Author's Name |
|
9th Author's Affiliation |
() |
10th Author's Name |
|
10th Author's Affiliation |
() |
11th Author's Name |
|
11th Author's Affiliation |
() |
12th Author's Name |
|
12th Author's Affiliation |
() |
13th Author's Name |
|
13th Author's Affiliation |
() |
14th Author's Name |
|
14th Author's Affiliation |
() |
15th Author's Name |
|
15th Author's Affiliation |
() |
16th Author's Name |
|
16th Author's Affiliation |
() |
17th Author's Name |
|
17th Author's Affiliation |
() |
18th Author's Name |
|
18th Author's Affiliation |
() |
19th Author's Name |
|
19th Author's Affiliation |
() |
20th Author's Name |
|
20th Author's Affiliation |
() |
Speaker |
Author-1 |
Date Time |
2021-03-03 10:00:00 |
Presentation Time |
25 minutes |
Registration for |
VLD |
Paper # |
VLD2020-67, HWS2020-42 |
Volume (vol) |
vol.120 |
Number (no) |
no.400(VLD), no.401(HWS) |
Page |
pp.1-6 |
#Pages |
6 |
Date of Issue |
2021-02-24 (VLD, HWS) |
|