AWS · AIF-C01
Transparent and explainable AI
AIF-C01 · Lesson 15 of 17 · 1:56:29
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Prepare for the AIF-C01 exam (AWS Certified AI Practitioner). In this lesson: transparent and explainable models, the transparency tools on AWS, from SageMaker Model Cards to AWS AI Service Cards, the tradeoffs between interpretability and performance, and between transparency and safety, and human-centered design. Includes three original practice questions.
Preview the topics
- AnyCompany must document its fraud model’s intended use, risk rating, training details, and evaluation results, in one place, for its auditors. Which feature fits?
- A deep model is three points more accurate than a decision tree on AnyCompany’s claim decisions, but the regulator requires a reason for each decision. What should the team do?
- AnyCompany wants customers to understand the AI’s claim decisions, and to challenge them. Which design choice fits human-centered explainable AI?