TechRxiv
TNNLS_2022___Contributor_Aware_Defenses_Against_Adversarial_Backdoor_Attacks.pdf (3.79 MB)
Download file

Contributor-Aware Defenses Against Adversarial Backdoor Attacks

Download (3.79 MB)
preprint
posted on 31.05.2022, 14:41 authored by Glenn DawsonGlenn Dawson, Muhammad Umer, Robi Polikar

Deep neural networks for image classifi cation are well-known to be vulnerable to adversarial attacks. One such attack that has garnered recent attention is the adversarial backdoor attack, which has demonstrated the capability to perform targeted misclassifi cation of specifi c examples. In particular, backdoor attacks attempt to force a model to learn spurious relations between backdoor trigger patterns and false labels. In response to this threat, numerous defensive measures have been proposed; however, defenses against backdoor attacks focus on backdoor pattern detection, which may be unreliable against novel or unexpected types of backdoor pattern designs. We introduce a novel re-contextualization of the adversarial setting, where the presence of an adversary implicitly admits the existence of multiple database contributors. Then, under the mild assumption of contributor awareness, it becomes possible to exploit this knowledge to defend against backdoor attacks by destroying the false label associations. We propose a contributor-aware universal defensive framework for learning in the presence of multiple, potentially adversarial data sources that utilizes semi-supervised ensembles and learning from crowds to fi lter the false labels produced by adversarial triggers. Importantly, this defensive strategy is agnostic to backdoor pattern design, as it functions without needing—or even attempting—to perform either adversary identifi cation or backdoor pattern detection during either training or inference. Our empirical studies demonstrate the robustness of the proposed framework against adversarial backdoor attacks from multiple simultaneous adversaries.

Funding

P200A180055

History

Email Address of Submitting Author

dawsong@rowan.edu

ORCID of Submitting Author

0000-0002-5877-5605

Submitting Author's Institution

Rowan University

Submitting Author's Country

United States of America