Clean-Label Backdoor Attacks
Deep neural networks have been recently demonstrated to be vulnerable to backdoor attacks. Specifically, by altering a small set of training examples, an adversary is able to install a backdoor that can be used during inference to fully control the model’s behavior. While the attack is very powerful, it crucially relies on the adversary being able to introduce arbitrary, often clearly mislabeled, inputs to the training set and can thus be detected even by fairly rudimentary data filtering. In this paper, we introduce a new approach to executing backdoor attacks, utilizing adversarial examples and GAN-generated data. The key feature is that the resulting poisoned inputs appear to be consistent with their label and thus seem benign even upon human inspection.
PDF Abstract