Adversarial Machine Learning Phases of Matter

Si Jiang, Sirui Lu, Dong-Ling Deng

We study the robustness of machine learning approaches to adversarial perturbations, with a focus on supervised learning scenarios. We find that typical phase classifiers based on…