Abstract: Deep Neural Networks (DNNs) have recently made significant strides in various fields; however, they are susceptible to adversarial examples—crafted inputs with imperceptible perturbations ...