Interpretable Machine Learning for Mitigating Feature-Driven Attacks | IEEE Journals & Magazine | IEEE Xplore