Formal Guarantees on the Robustness of a Classifier against Adversarial Manipulation

Open in new window