Attack Agnostic Adversarial Defense via Visual Imperceptible Bound
This addresses the problem of adversarial vulnerability in deep learning for security-critical applications, though it is incremental as it builds on existing defense concepts with a novel bound.
The paper tackles the lack of generalizability in adversarial defenses by proposing a Visual Imperceptible Bound (VIB) to create a robust defense model against both seen and unseen attacks, achieving improved robustness and maintaining or enhancing classification accuracy on clean datasets like MNIST, CIFAR-10, and Tiny ImageNet.
The high susceptibility of deep learning algorithms against structured and unstructured perturbations has motivated the development of efficient adversarial defense algorithms. However, the lack of generalizability of existing defense algorithms and the high variability in the performance of the attack algorithms for different databases raises several questions on the effectiveness of the defense algorithms. In this research, we aim to design a defense model that is robust within a certain bound against both seen and unseen adversarial attacks. This bound is related to the visual appearance of an image, and we termed it as \textit{Visual Imperceptible Bound (VIB)}. To compute this bound, we propose a novel method that uses the database characteristics. The VIB is further used to measure the effectiveness of attack algorithms. The performance of the proposed defense model is evaluated on the MNIST, CIFAR-10, and Tiny ImageNet databases on multiple attacks that include C\&W ($l_2$) and DeepFool. The proposed defense model is not only able to increase the robustness against several attacks but also retain or improve the classification accuracy on an original clean test set. The proposed algorithm is attack agnostic, i.e. it does not require any knowledge of the attack algorithm.