Your question is Robustness Against Adversarial Attacks. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
How do you ensure the robustness of a machine learning model against adversarial attacks?
Explain a practical evaluation strategy for identifying, measuring, and reducing performance degradation caused by adversarial inputs. Address attack generation, threat modeling, robust metrics, baseline comparisons, retraining, and monitoring after deployment. Discuss how your approach would change for white-box and black-box attackers, and how you would balance robustness with normal-data performance.