Use sensitive attributes to evaluate fairness for a small group.
Protect the data and prevent reidentification.
Choose two designs.
Limit purpose, access, and retention, use aggregation or anonymization, and share results at a level that does not identify small groups.
Detailed explanation
Fairness measurement and privacy are bounded together.
Fairness measurement and privacy are bounded together.
The result itself is less likely to expose people.
The result itself is less likely to expose people.
Evaluation data can become a privacy breach.
Evaluation data can become a privacy breach.
Harm cannot be detected.
Harm cannot be detected.
Try it yourself
An example you can run in a temporary verification environment.
AWS公式AIF-C01 Domain 4.1・5.2の公平性評価、プライバシー、データ最小化を確認する。Expected result
公平性を測るための属性データも保護対象として設計できる。Key points
- Purpose limitation
- Anonymization
- Reidentification
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
- Command output formatting can vary slightly by distribution or tool version.
- Run the example in a temporary directory or process when possible.
Foundation review
Read the scope first
Check whether the command acts on the current shell, a new process, an existing process, or a file.
Verify the observable result
Use the supplied command and compare the output with the expected result.