Detecting Adversarial Examples from Sensitivity Inconsistency of Spatial-Transform Domain
This addresses security vulnerabilities in AI systems by providing a more robust detection method for adversarial attacks, though it is incremental as it builds on existing detection paradigms.
The paper tackles the problem of detecting adversarial examples in deep neural networks by exploiting sensitivity inconsistency between original and transformed decision boundaries, achieving improved detection performance and superior generalization compared to state-of-the-art methods, especially for small adversarial perturbations.
Deep neural networks (DNNs) have been shown to be vulnerable against adversarial examples (AEs), which are maliciously designed to cause dramatic model output errors. In this work, we reveal that normal examples (NEs) are insensitive to the fluctuations occurring at the highly-curved region of the decision boundary, while AEs typically designed over one single domain (mostly spatial domain) exhibit exorbitant sensitivity on such fluctuations. This phenomenon motivates us to design another classifier (called dual classifier) with transformed decision boundary, which can be collaboratively used with the original classifier (called primal classifier) to detect AEs, by virtue of the sensitivity inconsistency. When comparing with the state-of-the-art algorithms based on Local Intrinsic Dimensionality (LID), Mahalanobis Distance (MD), and Feature Squeezing (FS), our proposed Sensitivity Inconsistency Detector (SID) achieves improved AE detection performance and superior generalization capabilities, especially in the challenging cases where the adversarial perturbation levels are small. Intensive experimental results on ResNet and VGG validate the superiority of the proposed SID.