4.1Standard
Do not use a protected attribute directly.
Another feature may strongly predict it.
Choose two evaluation practices.
× Incorrect
Inspect correlations and predictability for proxy features, and use protected attributes under proper controls to compare group outcomes.
Detailed explanation
CorrectTest feature correlations and predictability to identify proxy attributes
Indirect disparate impact can be found.
IncorrectIgnore group differences when overall performance is high
Averages can hide disparate impact.
IncorrectRemoving the protected field guarantees fairness
Proxy influence can remain.
CorrectUse protected attributes only for controlled evaluation and compare groups
Fairness measurement remains possible.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 4.1の公平性、代理属性、グループ評価を確認する。
Official references
What to confirm
直接・間接の属性影響を評価し、全体平均だけに依存しない。
Key points
- Proxy
- Correlation
- Group metrics
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Proxy attribute
A feature that can indirectly reveal a protected attribute through correlation.
Group evaluation
Measuring model performance or outcomes separately for defined populations.
Question IDAWS-AIF-330
EnvironmentAWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
Last technical review2026-08-20
Report an error or rights issue