Backdoor Pony: Evaluating Backdoor Attacks and Defenses in Different Domains

Mercier, Arthur; Smolin, Nikita; Sihlovec, Oliver; Koffas, Stefanos; Picek, Stjepan

doi:10.2139/ssrn.4330225

Cited by 1 publication

(1 citation statement)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since then, an increasing number of studies [60], [61], [62], [63], [58], [64], [65], [66], [21], [67], [68], and [44], [69] have been dedicated to the subject, significantly increasing the class of available attacks and the application scenarios potentially targeted by backdooring attempts. The proposed attacks differ on the basis of the event triggering the backdoor at test time, malicious behavior induced by the activation of the backdoor, stealthiness of the procedure used to inject the backdoor, modality through which the attacker interferes with the training process, and knowledge that the attacker has about the attacked network.…”

Section: Introductionmentioning

confidence: 99%

Backdoor Attacks to Deep Neural Networks: A Survey of the Literature, Challenges, and Future Research Directions

Mengara,

Avila,

Falk

2024

IEEE Access

View full text Add to dashboard Cite

Deep neural network (DNN) classifiers are potent instruments that can be used in various security-sensitive applications. Nonetheless, they are vulnerable to certain attacks that impede or distort their learning process. For example, backdoor attacks involve polluting the DNN learning set with a few samples from one or more source classes, which are then labelled as a target class set by an attacker. Even if the DNN is trained on clean samples with no backdoors, this attack will still be successful if a backdoor pattern exists in the training data. Backdoor attacks are difficult to spot and can be used to make the DNN behave maliciously depending on the target selected by the attacker. In this study, we survey the literature and highlight the latest advances in backdoor attack strategies and defense mechanisms. We finalize the discussion on challenges and open issues, as well as future research opportunities.

show abstract