Binary calibration on 2184 binary experiments (test)
-0.0034Brier Score GapQuadratic scaling
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Quadratic scaling2025.11 | -0.0034 | -0.0181 | |
| Affine scaling2025.11 | -0.0026 | -0.0154 | |
| Isotonic regression2025.11 | -0.0025 | 0.0909 | |
| TS (probmetrics)implementation=probmetrics2025.11 | -0.0018 | -0.0167 | |
| TS (ours)implementation=ours2025.11 | -0.0018 | -0.0142 | |
| TS (Guo et al.)implementation=Guo et al.2025.11 | 0.0063 | 0.0043 |