Few-shot Domain Generalization on Meta-Dataset ImageNet-only v1 (train)
56.66Accuracy (ImageNet)GAP + Proto-MAML
Evaluation results
| Method | Links | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GAP + Proto-MAMLBackbone=ResNet-18, Training Setting=ImageNet only2023.04 | 56.66 | 77.57 | 68.5 | 73.51 | 71.42 | 65.36 | 38.57 | 86.79 | 66.9 | 46.76 | 93.97 | 74.34 | 65.23 | 56.66 | 66.26 | 65.3 | 69.28 | 68.06 | |
| Approximate GAP + Proto-MAMLBackbone=ResNet-18, Training Setting=ImageNet only2023.04 | 56.65 | 76.3 | 67.61 | 73.29 | 71.01 | 63.71 | 38.3 | 86.73 | 66.44 | 46.54 | 92.27 | 75.96 | 65.94 | 56.65 | 65.65 | 64.66 | 68.39 | 67.35 | |
| GAP + fo-MAMLBackbone=ResNet-18, Training Setting=ImageNet only2023.04 | 56.33 | 77.04 | 68.03 | 72.01 | 71.82 | 64.36 | 37.46 | 85.31 | 67.89 | 46.87 | 93.53 | 74.54 | 63.23 | 56.33 | 65.64 | 64.71 | 70.04 | 67.89 | |
| Approximate GAP + fo-MAMLBackbone=ResNet-18, Training Setting=ImageNet only2023.04 | 55.34 | 75.17 | 67.4 | 72.35 | 71.12 | 63.49 | 37.19 | 84.44 | 66.82 | 45.99 | 93.46 | 74.05 | 62.66 | 55.34 | 64.89 | 63.93 | 69.56 | 67.25 | |
| ALFA + Proto-MAMLBackbone=ResNet-18, Training Setting=ImageNet only2023.04 | 52.8 | 61.87 | 63.43 | 69.75 | 70.78 | 59.17 | 41.49 | 85.96 | 60.78 | 48.11 | — | — | — | 52.8 | 62.37 | 61.41 | — | — | |
| ALFA + fo-MAMLBackbone=ResNet-18, Training Setting=ImageNet only2023.04 | 51.09 | 67.89 | 66.34 | 67.67 | 65.34 | 60.53 | 37.41 | 84.28 | 60.86 | 40.05 | — | — | — | 51.09 | 61.15 | 60.15 | — | — | |
| Proto-MAMLBackbone=ResNet-18, Training Setting=ImageNet only2023.04 | 49.53 | 63.37 | 55.95 | 68.66 | 66.49 | 51.52 | 39.96 | 87.15 | 48.83 | 43.74 | — | — | — | 49.53 | 58.41 | 57.52 | — | — | |
| fo-MAMLBackbone=ResNet-18, Training Setting=ImageNet only2023.04 | 45.51 | 55.55 | 56.24 | 63.61 | 68.04 | 43.96 | 32.1 | 81.74 | 50.93 | 35.3 | — | — | — | 45.51 | 54.16 | 53.3 | — | — |