Strategic safeguarding: A game theoretic approach for analyzing attacker-defender behavior in DNN backdoors

Abstract Deep neural networks (DNNs) are fundamental to modern applications like face recognition and autonomous driving. However, their security is a significant concern due to various integrity risks, such as backdoor attacks. In these attacks, compromised training data introduce malicious behavio...

Full description

Bibliographic Details

Published in:	EURASIP Journal on Information Security
Main Authors:	Kassem Kallas, Quentin Le Roux, Wassim Hamidouche, Teddy Furon
Format:	Article
Language:	English
Published:	SpringerOpen 2024-10-01
Subjects:	Deep neural networks CNN Backdoor attacks Backdoor defenses Game theory Zero-sum game
Online Access:	https://doi.org/10.1186/s13635-024-00180-5