When Explainability Meets Adversarial Learning: Detecting Adversarial Examples using SHAP Signatures

When Explainability Meets Adversarial Learning: Detecting Adversarial Examples using SHAP Signatures Gil Fidel Ron Bitton Asaf Shabtai Department of Software and Information Systems Engineering Ben-Gurion University of the Negev Abstract --State-of-the-art deep neural networks (DNNs) are highly effective in solving many complex real-world problems. However, these models are vulnerable to adversarial perturbation attacks, and despite the plethora of research in this domain, to this day, adversaries still have the upper hand in the cat and mouse game of adversarial example generation methods vs. detection and prevention methods. In this research, we present a novel detection method that uses Shapley Additive Explanations (SHAP) values computed for the internal layers of a DNN classifier to discriminate between normal and adversarial inputs. We evaluate our method by building an extensive dataset of adversarial examples over the popular CIF AR-10 and MNIST datasets, and training a neural network-based detector to distinguish between normal and adversarial inputs. We evaluate our detector against adversarial examples generated by diverse state-of-the-art attacks and demonstrate its high detection accuracy and strong generalization ability to adversarial inputs generated with different attack methods. I NTRODUCTION In recent years, deep neural network (DNN) learning algorithms have been widely used to solve a variety of complex problems. Their greatest impact has been seen in fields such as image classification, object recognition, natural language processing, and malware detection.