Abstract: Numerous adversarial attack methods have been developed to generate imperceptible image perturbations that cause erroneous predictions in state-of-the-art machine learning (ML) models, ...