Conformal Prediction Evaluation on CIFAR100
84.73Top-1 AccuracyLabel Distance
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Label DistanceConformal domain=Probability space, Distance metric=Cosine distance, Model=ResNet-502026.05 | 84.73 | 97.11 | 5.568 | |
| Margin DistanceConformal domain=Logit space, Distance metric=Cosine distance, Model=ResNet-502026.05 | 84.73 | 97.11 | 5.352 | |
| Mean DistanceConformal domain=Feature space, Distance metric=Cosine distance, Model=ResNet-502026.05 | 84.73 | 97.11 | 2.785 | |
| APS (U=0.001)Conformal domain=Probability space, Distance metric=N/A, Model=ResNet-502026.05 | 84.73 | 97.11 | 6.709 | |
| RAPS (λ = 0.2, k = 1, U=0.001)Conformal domain=Probability space, Distance metric=N/A, Model=ResNet-502026.05 | 84.73 | 97.11 | 4.922 | |
| SAPS (λ = 0.2, U=0.001)Conformal domain=Probability space, Distance metric=N/A, Model=ResNet-502026.05 | 84.73 | 97.11 | 4.754 | |
| Fast GradientConformal domain=Feature space, Distance metric=Euclidean distance, Model=ResNet-502026.05 | 84.73 | 97.11 | 13.11 |