AWS Certification Exam AI Practitioner (Practice Questions) set17
通常
試験時間: 00:00問題時間: 00:00
問題一覧
問題 53 / 65
Which of the following is the correct explanation regarding countermeasures against adversarial attacks and sensor errors for traffic control AI?
①Interpretable models are absolutely robust to adversarial attacks.
②Sensor errors can be ignored without affecting system safety.
③A black-box model is completely unaffected by external interference.
④All models require robustness evaluation and defenses, and implementing countermeasures requires additional design and verification.
⑤Adversarial attacks cannot occur in real-world traffic systems.