Black-box Adversarial Example Generation with Normalizing Flows
This addresses the issue of adversarial attacks for improving model robustness, but it is incremental as it builds on existing black-box methods with a new technique.
The paper tackled the problem of adversarial vulnerability in deep neural network classifiers by proposing a black-box adversarial attack using normalizing flows, which generates adversaries that closely resemble the original data and demonstrated competitive performance against existing methods.
Deep neural network classifiers suffer from adversarial vulnerability: well-crafted, unnoticeable changes to the input data can affect the classifier decision. In this regard, the study of powerful adversarial attacks can help shed light on sources of this malicious behavior. In this paper, we propose a novel black-box adversarial attack using normalizing flows. We show how an adversary can be found by searching over a pre-trained flow-based model base distribution. This way, we can generate adversaries that resemble the original data closely as the perturbations are in the shape of the data. We then demonstrate the competitive performance of the proposed approach against well-known black-box adversarial attack methods.