OOD Detection on MNIST (In-dist) vs OMNIGLOT (OOD) (test)
99.98AUCBNN-ARHT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| BNN-ARHTIn-Distrib.=MNIST, Architecture=LeNet [22]2023.10 | 99.98 | 99.98 | |
| DPNIn-Distrib.=MNIST, Architecture=LeNet [22]2023.10 | 99.96 | 99.96 | |
| MC DropoutIn-Distrib.=MNIST, Architecture=LeNet [22]2023.10 | 99.85 | 99.88 | |
| Deep EnsemblesIn-Distrib.=MNIST, Architecture=LeNet [22]2023.10 | 99.7 | 91.08 | |
| DetectronIn-Distrib.=MNIST, Architecture=LeNet [22]2023.10 | 95.71 | 85 | |
| Kendall and GalIn-Distrib.=MNIST, Architecture=LeNet [22]2023.10 | 94.11 | 93.4 | |
| EDLIn-Distrib.=MNIST, Architecture=LeNet [22]2023.10 | 72.61 | 81.42 |