Out-of-Distribution Detection on FashionMNIST vs NotMNIST
0.9726AUROCFSVI
Evaluation results
| Method | Links | |
|---|---|---|
| FSVIArchitecture=four-layer CNN, Ensemble=false, Context distribution (p_xc)=KMNIST2023.12 | 0.9726 | |
| FSVI EnsembleArchitecture=four-layer CNN, Ensemble=true, Context distribution (p_xc)=random monochrome2023.12 | 0.9695 | |
| FSVIArchitecture=four-layer CNN, Ensemble=false, Context distribution (p_xc)=random monochrome2023.12 | 0.9502 | |
| DUQArchitecture=four-layer CNN, Ensemble=false2023.12 | 0.946 | |
| MFVIArchitecture=four-layer CNN, Ensemble=false2023.12 | 0.8888 | |
| Deep EnsembleArchitecture=four-layer CNN, Ensemble=true2023.12 | 0.8317 | |
| MFVI (radial)Architecture=four-layer CNN, Ensemble=false2023.12 | 0.8211 | |
| MFVI (tempered)Architecture=four-layer CNN, Ensemble=false2023.12 | 0.8078 | |
| SWAGArchitecture=four-layer CNN, Ensemble=false2023.12 | 0.8031 | |
| MC DROPOUTArchitecture=four-layer CNN, Ensemble=false2023.12 | 0.8002 | |
| MAPArchitecture=four-layer CNN, Ensemble=false2023.12 | 0.7485 |