Post-Training Detection of Backdoor Attacks for Two-Class and Multi-Attack Scenarios. (arXiv:2201.08474v1 [cs.CR])

Backdoor attacks (BAs) are an emerging threat to deep neural network
classifiers. A victim classifier will predict to an attacker-desired target
class whenever a test sample is embedded with the same backdoor pattern (BP)
that was used to poison the classifier's training set. Detecting whether a
classifier is backdoor attacked is not easy in practice, especially when the
defender is, e.g., a downstream user without access to the classifier's
training set. This challenge is addressed here by a reverse-engineering defense …

arxiv attacks detection training