Article (Scientific journals)
Evolutionary algorithms deceive humans and machines at image classification: An extended proof of concept on two scenarios
CHITIC, Ioana Raluca; LEPREVOST, Franck; Bernard, Nicolas
2020In Journal of Information and Telecommunication
Peer Reviewed verified by ORBi
 

Files


Full Text
A_proof_of_concept_on_two_scenarii_to_deceive_humans_and_machines_at_image_classification_with_evolutionary_algorithms.pdf
Publisher postprint (2.06 MB)
Download

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
Neural Networks; Evolutionary Algorithms; Adversarial Attacks
Abstract :
[en] The range of applications of Neural Networks encompasses image classification. However, Neural Networks are vulnerable to attacks, and may misclassify adversarial images, leading to potentially disastrous consequences. Pursuing some of our previous work, we provide an extended proof of concept of a black-box, targeted, non-parametric attack using evolutionary algorithms to fool both Neural Networks and humans at the task of image classification. Our feasibility study is performed on VGG-16 trained on CIFAR-10. For any category cA of CIFAR-10, one chooses an image A classified by VGG-16 as belonging to cA. From there, two scenarios are addressed. In the first scenario, a target category ct≠cA is fixed a priori. We construct an evolutionary algorithm that evolves A to a modified image that VGG-16 classifies as belonging to ct. In the second scenario, we construct another evolutionary algorithm that evolves A to a modified image that VGG-16 is unable to classify. In both scenarios, the obtained adversarial images remain so close to the original one that a human would likely classify them as still belonging to cA.
Disciplines :
Computer science
Author, co-author :
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)
Bernard, Nicolas;  independent
External co-authors :
no
Language :
English
Title :
Evolutionary algorithms deceive humans and machines at image classification: An extended proof of concept on two scenarios
Publication date :
10 October 2020
Journal title :
Journal of Information and Telecommunication
ISSN :
2475-1839
eISSN :
2475-1847
Publisher :
Taylor & Francis Group, United Kingdom
Peer reviewed :
Peer Reviewed verified by ORBi
Available on ORBilu :
since 10 November 2020

Statistics


Number of views
129 (17 by Unilu)
Number of downloads
61 (5 by Unilu)

Scopus citations®
 
5
Scopus citations®
without self-citations
0
OpenCitations
 
3
WoS citations
 
7

Bibliography


Similar publications



Contact ORBilu