Score distributions and error rates differ by group.
A single threshold may not have the same effect.
Choose two evaluations.
Compare subgroup precision, recall, and selection rate by threshold and agree on fairness, legal, and operational requirements with stakeholders.
Detailed explanation
Threshold impact is visible.
Threshold impact is visible.
Trade-offs are decided transparently.
Trade-offs are decided transparently.
Score distributions and errors can differ.
Score distributions and errors can differ.
Harm and remedy are ignored.
Harm and remedy are ignored.
Try it yourself
An example you can run in a temporary verification environment.
AWS公式AIF-C01 Domain 4.1の公平性、閾値、バイアス評価を確認する。Expected result
単一の閾値や平均性能だけで公平性を判断しない理由を説明できる。Key points
- Threshold
- Group metrics
- Requirement
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.