Conformal Prediction on ImageNet ILSVRC2012 (test)
2.312Avg Prediction Set SizeRAPS+TTA-Learned
Evaluation results
| Method | Links | |
|---|---|---|
| RAPS+TTA-LearnedAlpha=0.1, Augmentation Policy=Expanded Augmentation Policy2025.05 | 2.312 | |
| RAPS+TTA-AvgAlpha=0.1, Augmentation Policy=Simple Augmentation Policy2025.05 | 2.327 | |
| RAPS+TTA-LearnedAlpha=0.1, Augmentation Policy=Simple Augmentation Policy2025.05 | 2.362 | |
| RAPS+TTA-AvgAlpha=0.1, Augmentation Policy=Expanded Augmentation Policy2025.05 | 2.47 | |
| RAPSAlpha=0.1, Augmentation Policy=Expanded Augmentation Policy2025.05 | 2.548 | |
| RAPSAlpha=0.1, Augmentation Policy=Simple Augmentation Policy2025.05 | 2.548 | |
| RAPS+TTA-LearnedAlpha=0.05, Augmentation Policy=Expanded Augmentation Policy2025.05 | 4.889 | |
| RAPS+TTA-AvgAlpha=0.05, Augmentation Policy=Simple Augmentation Policy2025.05 | 4.908 | |
| RAPS+TTA-LearnedAlpha=0.05, Augmentation Policy=Simple Augmentation Policy2025.05 | 5.04 | |
| RAPS+TTA-AvgAlpha=0.05, Augmentation Policy=Expanded Augmentation Policy2025.05 | 5.318 | |
| RAPSAlpha=0.05, Augmentation Policy=Expanded Augmentation Policy2025.05 | 5.637 | |
| RAPSAlpha=0.05, Augmentation Policy=Simple Augmentation Policy2025.05 | 5.637 | |
| RAPS+TTA-LearnedAlpha=0.01, Augmentation Policy=Expanded Augmentation Policy2025.05 | 31.248 | |
| RAPS+TTA-AvgAlpha=0.01, Augmentation Policy=Simple Augmentation Policy2025.05 | 31.681 | |
| RAPS+TTA-LearnedAlpha=0.01, Augmentation Policy=Simple Augmentation Policy2025.05 | 32.702 | |
| RAPS+TTA-AvgAlpha=0.01, Augmentation Policy=Expanded Augmentation Policy2025.05 | 35.6 | |
| RAPSAlpha=0.01, Augmentation Policy=Expanded Augmentation Policy2025.05 | 37.751 | |
| RAPSAlpha=0.01, Augmentation Policy=Simple Augmentation Policy2025.05 | 37.751 |