Out-of-Distribution Detection on CIFAR10 (In-distribution) vs SVHN (OOD) (test)
91.61AUPRBNN-ARHT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| BNN-ARHTBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 91.61 | 82.01 | |
| BNN-ARHTIn-Distrib.=CIFAR10, Architecture=LeNet [22]2023.10 | 91.61 | 82.01 | |
| DetectronBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 90 | 76.01 | |
| DetectronIn-Distrib.=CIFAR10, Architecture=LeNet [22]2023.10 | 90 | 76.01 | |
| MC DropoutBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 84.35 | 78.09 | |
| MC DropoutIn-Distrib.=CIFAR10, Architecture=LeNet [22]2023.10 | 84.35 | 78.09 | |
| EDLBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 83.74 | 69.57 | |
| EDLIn-Distrib.=CIFAR10, Architecture=LeNet [22]2023.10 | 83.74 | 69.57 | |
| I-EDLBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 83.26 | — | |
| Deep EnsemblesBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 82.62 | 76.15 | |
| Deep EnsemblesIn-Distrib.=CIFAR10, Architecture=LeNet [22]2023.10 | 82.62 | 76.15 | |
| DPNBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 77.76 | 57.48 | |
| DPNIn-Distrib.=CIFAR10, Architecture=LeNet [22]2023.10 | 77.76 | 57.48 | |
| Kendall and GalBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 71.44 | 67.4 | |
| Kendall and GalIn-Distrib.=CIFAR10, Architecture=LeNet [22]2023.10 | 71.44 | 67.4 | |
| PostNetBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 69.3 | 76.04 | |
| PostNetIn-Distrib.=CIFAR10, Architecture=LeNet [22]2023.10 | 69.3 | 76.04 | |
| RKL-PNBackbone=LeNet, In-Distribution Dataset=CIFAR102023.10 | 61.41 | 57.89 |