Information: Join today and make your research activities more affordable! Technical workshop participation fees and annual registration fees are available at member rates.
Notice: [Important] Announcement of Changes to Registration Fee Payment and Manuscript Upload Procedures for IEICE Technical Meetings[Type 1 Only]
IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2024-03-01 10:10
Fault Detectable Convolutional Neural Network Circuits With Dual Modular Redundancy Based on Mixed-precision Quantization
Yamato Saikawa, Yuta Owada, Yoichi Tomioka, Hiroshi Saito, Yukihide Kohira (UoA) VLD2023-122 HWS2023-82 ICD2023-111
Abstract (in Japanese) (See Japanese page) 
(in English) In safety-critical edge AI systems, circuit failures caused by aging or cosmic ray can lead to serious accidents. Dual Modular Redundancy (DMR) is a conventional technique for fault detection. However, it significantly increases the area and power consumption of convolutional neural network (CNN) circuits, making it difficult to use DMR in edge devices with strict size and power constraints. In this paper, we propose an approximate DMR that combines convolutional layers with different quantization bit widths based on mixed-precision quantization and evaluate its fault detection performance, resource usage, and power consumption when implemented in an FPGA device. We apply the approximate DMR to ResNet-20 and show that high fault detection performance can be achieved. Moreover, we show that a 4-bit quantized convolutional layer, including a quantization layer of input feature map, can reduce the number of LUTs and power consumption by 52.9% and 53.3%, respectively, compared to 8-bit quantized convolutional layer. We also show that the proposed DMR circuit of three residual blocks reduces the number of LUTs and power consumption by 27.0% and 19.1%, respectively.
Keyword (in Japanese) (See Japanese page) 
(in English) Convolutional neural network / Fault-tolerance / Dual Modular Redundancy / Quantization / / / /  
Reference Info. IEICE Tech. Rep., vol. 123, no. 390, VLD2023-122, pp. 119-124, Feb. 2024.
Paper # VLD2023-122 
Date of Issue 2024-02-21 (VLD, HWS, ICD) 
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 VLD2023-122 HWS2023-82 ICD2023-111

Conference Information
Committee VLD HWS ICD  
Conference Date 2024-02-28 - 2024-03-02 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To VLD 
Conference Code 2024-02-VLD-HWS-ICD 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Fault Detectable Convolutional Neural Network Circuits With Dual Modular Redundancy Based on Mixed-precision Quantization 
Sub Title (in English)  
Keyword(1) Convolutional neural network  
Keyword(2) Fault-tolerance  
Keyword(3) Dual Modular Redundancy  
Keyword(4) Quantization  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Yamato Saikawa  
1st Author's Affiliation University of Aizu (UoA)
2nd Author's Name Yuta Owada  
2nd Author's Affiliation University of Aizu (UoA)
3rd Author's Name Yoichi Tomioka  
3rd Author's Affiliation University of Aizu (UoA)
4th Author's Name Hiroshi Saito  
4th Author's Affiliation University of Aizu (UoA)
5th Author's Name Yukihide Kohira  
5th Author's Affiliation University of Aizu (UoA)
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 ()
21st Author's Name  
21st Author's Affiliation ()
22nd Author's Name  
22nd Author's Affiliation ()
23rd Author's Name  
23rd Author's Affiliation ()
24th Author's Name  
24th Author's Affiliation ()
25th Author's Name  
25th Author's Affiliation ()
26th Author's Name / /
26th Author's Affiliation ()
()
27th Author's Name / /
27th Author's Affiliation ()
()
28th Author's Name / /
28th Author's Affiliation ()
()
29th Author's Name / /
29th Author's Affiliation ()
()
30th Author's Name / /
30th Author's Affiliation ()
()
31st Author's Name / /
31st Author's Affiliation ()
()
32nd Author's Name / /
32nd Author's Affiliation ()
()
33rd Author's Name / /
33rd Author's Affiliation ()
()
34th Author's Name / /
34th Author's Affiliation ()
()
35th Author's Name / /
35th Author's Affiliation ()
()
36th Author's Name / /
36th Author's Affiliation ()
()
Speaker Author-1 
Date Time 2024-03-01 10:10:00 
Presentation Time 25 minutes 
Registration for VLD 
Paper # VLD2023-122, HWS2023-82, ICD2023-111 
Volume (vol) vol.123 
Number (no) no.390(VLD), no.391(HWS), no.392(ICD) 
Page pp.119-124 
#Pages 6 
Date of Issue 2024-02-21 (VLD, HWS, ICD) 


[Return to Top Page]

[Return to IEICE Web Page]

(Line Height for English Pages: 1)

The Institute of Electronics, Information and Communication Engineers (IEICE), Japan