Selective Prediction: When Models Should Abstain
A practical guide to risk-coverage trade-offs: use uncertainty to decide which predictions to accept, review, or defer.
What it covers
- Risk and coverage
- Build the curve, not one threshold
- Read the curve as a system decision
- Separate ranking from calibration
- Test the failure modes
- Respect dependence
- Inspect subgroups
- Include a random baseline
- Account for review capacity
- A practical deployment contract
- References
Topics covered so far
- Uncertainty Quantification
- Applied
- Model Evaluation
- Selective Prediction
The library grows deliberately. A piece is published when the underlying work is finished and its limitations are known, not on a schedule.