Open Set Domain Generalization on Digits-DG SVHN (test)
80.06AccuracyCIRL
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CIRLBackbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 80.06 | 58.73 | 56.88 | |
| MEDIC++Backbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 77.31 | 61.55 | 59.9 | |
| SWADBackbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 76.08 | 59.18 | 56.25 | |
| MEDICBackbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 76 | 58.77 | 57.6 | |
| FishBackbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 74.42 | 49.94 | 52.89 | |
| MLDGBackbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 73.64 | 53.54 | 53.64 | |
| OpenMaxBackbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 72 | 49.46 | — | |
| ARPLBackbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 71.78 | 56.98 | 53.63 | |
| ERMBackbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 71.08 | 54.37 | 49.86 | |
| MixStyleBackbone=ConvNet, Known:Unknown Ratio=6:42026.06 | 69.28 | 53.24 | 48.43 |