Adversarial Defenses

My notes on certifiable robustness of ML models.