4.1Standard
One group has a high false-rejection rate.
Find the cause and reduce impact.
Choose two responses.
× Incorrect
Separate causes across labels, features, sampling, thresholds, and human review, then reevaluate group performance and business impact with stakeholders.
Detailed explanation
IncorrectAssume lowering overall accuracy always fixes the gap
Effectiveness must be measured.
CorrectReevaluate group quality and business impact and stage mitigations with approval
Side effects are checked.
CorrectDecompose causes across labels, features, sampling, thresholds, and human review
The intervention point can be identified.
IncorrectRemove the affected group from training
Impact is hidden or worsened.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 4.1の公平性、緩和策、影響評価を確認する。
Official references
What to confirm
公平性の差を原因段階へ分解し、検証付きで対策を導入できる。
Key points
- Fairness
- Mitigation
- Side effect
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
Fairness mitigation
A change intended to reduce unjustified performance differences between groups.
Error disparity
A difference in error rates or error types across defined groups.
Question IDAWS-AIF-403
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