1.2Standard
Production feature units and distributions changed.
Investigate model degradation.
Choose two checks.
× Incorrect
Compare distributions, units, ranges, missingness, and categories with training data, and correlate changes with model version, performance, and label delay.
Detailed explanation
CorrectCorrelate changes with model version, performance, and label delay over time
Root causes can be separated.
CorrectCompare distributions, units, ranges, missingness, and categories by feature
Input changes can be identified.
IncorrectIgnore a unit change when the distribution looks similar
The feature meaning can change.
IncorrectRetrain every model after one feature changes
Impact has not been established.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 1.2の特徴量ドリフト、データ品質、モデル監視を確認する。
Official references
What to confirm
特徴量の分布・意味の変化と性能劣化を関連付けて調査できる。
Key points
- Feature drift
- Units
- Time series
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
Feature drift
A change in the distribution or meaning of a model input feature over time.
Data contract
An agreement describing input fields, types, units, ranges, missingness, and update timing.
Question IDAWS-AIF-393
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