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 2022-06-16 14:40
Adversarial Robustness of Secret Key-Based Defenses against AutoAttack
Miki Tanaka, April Pyone MaungMaung (Tokyo Metro Univ.), Isao Echizen (NII), Hitoshi Kiya (Tokyo Metro Univ.) CAS2022-7 VLD2022-7 SIP2022-38 MSS2022-7
Abstract (in Japanese) (See Japanese page) 
(in English) Deep neural network (DNN) models are well-known to easily misclassify prediction results by using input images with small perturbations, called adversarial examples, so investigating countermeasures for adversarial examples is an urgent issue. In this paper, the secret key-based defense that we proposed is evaluated in terms of robustness against adversarial examples in accordance with a benchmark attack method, called AutoAttack. In addition, we propose a detection method of adversarial examples to be combined with the secret key-based defense. In an experiment, the secret key-based classification model is confirmed that it is not robust enough against a black box attack, and the combined use of the key-based defense and the proposed detector outperforms the latest benchmark.
Keyword (in Japanese) (See Japanese page) 
(in English) Adversarial example / Machine learning / Deep learning / Adversarial detection / / / /  
Reference Info. IEICE Tech. Rep., vol. 122, no. 77, SIP2022-38, pp. 34-39, June 2022.
Paper # SIP2022-38 
Date of Issue 2022-06-09 (CAS, VLD, SIP, MSS) 
ISSN Online edition: ISSN 2432-6380
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 CAS2022-7 VLD2022-7 SIP2022-38 MSS2022-7

Conference Information
Committee CAS SIP VLD MSS  
Conference Date 2022-06-16 - 2022-06-17 
Place (in Japanese) (See Japanese page) 
Place (in English) Hachinohe Institute of Technology 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To SIP 
Conference Code 2022-06-CAS-SIP-VLD-MSS 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Adversarial Robustness of Secret Key-Based Defenses against AutoAttack 
Sub Title (in English)  
Keyword(1) Adversarial example  
Keyword(2) Machine learning  
Keyword(3) Deep learning  
Keyword(4) Adversarial detection  
1st Author's Name Miki Tanaka  
1st Author's Affiliation Tokyo Metropolitan University (Tokyo Metro Univ.)
2nd Author's Name April Pyone MaungMaung  
2nd Author's Affiliation Tokyo Metropolitan University (Tokyo Metro Univ.)
3rd Author's Name Isao Echizen  
3rd Author's Affiliation National Institute of Informatics (NII)
4th Author's Name Hitoshi Kiya  
4th Author's Affiliation Tokyo Metropolitan University (Tokyo Metro Univ.)
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 2022-06-16 14:40:00 
Presentation Time 25 minutes 
Registration for SIP 
Paper # CAS2022-7, VLD2022-7, SIP2022-38, MSS2022-7 
Volume (vol) vol.122 
Number (no) no.75(CAS), no.76(VLD), no.77(SIP), no.78(MSS) 
Page pp.34-39 
Date of Issue 2022-06-09 (CAS, VLD, SIP, MSS) 

[Return to Top Page]

[Return to IEICE Web Page]

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