Hima Lakkaraju, assistant professor at Harvard University, explains how to monitor the adversarial robustness of deployed ML models.
