Article (Scientific journals)
One evolutionary algorithm deceives humans and ten convolutional neural networks trained on ImageNet at image recognition
TOPAL, Ali Osman; CHITIC, Ioana Raluca; LEPREVOST, Franck
2023In Applied Soft Computing, 143, p. 110397
Peer Reviewed verified by ORBi
 

Files


Full Text
1-s2.0-S1568494623004155-main.pdf
Publisher postprint (7.38 MB)
Download

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
Adversarial attacks; Black-box attacks; Convolutional neural network; Evolutionary algorithm; Image classification
Abstract :
[en] Convolutional neural networks (CNNs) are widely used in computer vision, but can be deceived by carefully crafted adversarial images. In this paper, we propose an evolutionary algorithm (EA) based adversarial attack against CNNs trained on ImageNet. Our EA-based attack aims to generate adversarial images that not only achieve a high confidence probability of being classified into the target category (at least 75%), but also appear indistinguishable to the human eye in a black-box setting. These constraints are implemented to simulate a realistic adversarial attack scenario. Our attack has been thoroughly evaluated on 10 CNNs in various attack scenarios, including high-confidence targeted, good-enough targeted, and untargeted. Furthermore, we have compared our attack favorably against other well-known white-box and black-box attacks. The experimental results revealed that the proposed EA-based attack is superior or on par with its competitors in terms of the success rate and the visual quality of the adversarial images produced.
Research center :
ULHPC - University of Luxembourg: High Performance Computing
Disciplines :
Computer science
Author, co-author :
TOPAL, Ali Osman ;  University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS)
CHITIC, Ioana Raluca ;  University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS)
LEPREVOST, Franck ;  University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS)
External co-authors :
no
Language :
English
Title :
One evolutionary algorithm deceives humans and ten convolutional neural networks trained on ImageNet at image recognition
Publication date :
11 May 2023
Journal title :
Applied Soft Computing
ISSN :
1568-4946
eISSN :
1872-9681
Publisher :
Elsevier
Volume :
143
Pages :
110397
Peer reviewed :
Peer Reviewed verified by ORBi
Focus Area :
Computational Sciences
Available on ORBilu :
since 22 June 2023

Statistics


Number of views
96 (8 by Unilu)
Number of downloads
77 (2 by Unilu)

Scopus citations®
 
13
Scopus citations®
without self-citations
10

Bibliography


Similar publications



Contact ORBilu