Image Classification on Digit-5 feature shift (test)
72.86Accuracy (R=1)LSS
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LSSBackbone=ResNet-18, Pre-trained=ImageNet, Memory=4x2024.10 | 72.86 | 92.97 | |
| DiWABackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 61.54 | 88.83 | |
| SoupsBackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 59.71 | 87.07 | |
| SWADBackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 57.02 | 86.84 | |
| SWABackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 54.13 | 85.33 | |
| MOONBackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 50.11 | 83.02 | |
| FedBABUBackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 48.02 | 83.2 | |
| FedRepBackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 47.43 | 82.02 | |
| FedAvgBackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 46.36 | 80.48 | |
| FedBNBackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 46.02 | 81.42 | |
| FedProxBackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 44.01 | 77.83 | |
| FedFomoBackbone=ResNet-18, Pre-trained=ImageNet2024.10 | 41.87 | 76.21 |