Medical Image Classification on BreaKHis
85.97AccMH-pFLGB
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MH-pFLGBModel Architecture=Heterogeneous (ResNet17, 11, 8, 5)2024.06 | 85.97 | 82.47 | — | — | — | |
| KT-pFLModel Architecture=Heterogeneous (ResNet17, 11, 8, 5)2024.06 | 79.25 | 74.22 | — | — | — | |
| pFedDFModel Architecture=Heterogeneous (ResNet17, 11, 8, 5)2024.06 | 78.92 | 73.66 | — | — | — | |
| Only Local TrainingModel Architecture=Heterogeneous (ResNet17, 11, 8, 5)2024.06 | 76.03 | 68.75 | — | — | — | |
| FedRepModel Architecture=Unified (ResNet17)2024.06 | 75.78 | 68.97 | — | — | — | |
| DS-PFLModel Architecture=Heterogeneous (ResNet17, 11, 8, 5)2024.06 | 75.71 | 70.16 | — | — | — | |
| FedDFModel Architecture=Heterogeneous (ResNet17, 11, 8, 5)2024.06 | 75.62 | 69.63 | — | — | — | |
| FedMDModel Architecture=Heterogeneous (ResNet17, 11, 8, 5)2024.06 | 75.34 | 68.11 | — | — | — | |
| SCAFFOLD+FTModel Architecture=Unified (ResNet17), Fine-tuned=true2024.06 | 74.16 | 67.04 | — | — | — | |
| FedProx+FTModel Architecture=Unified (ResNet17), Fine-tuned=true2024.06 | 73.75 | 66.27 | — | — | — | |
| FedAvg+FTModel Architecture=Unified (ResNet17), Fine-tuned=true2024.06 | 73.59 | 66.08 | — | — | — | |
| SCAFFOLDModel Architecture=Unified (ResNet17)2024.06 | 70.33 | 61.85 | — | — | — | |
| FedProxModel Architecture=Unified (ResNet17)2024.06 | 69.98 | 61.5 | — | — | — | |
| FedAvgModel Architecture=Unified (ResNet17)2024.06 | 69.95 | 61.36 | — | — | — | |
| APFLModel Architecture=Unified (ResNet17)2024.06 | 69.49 | 60.35 | — | — | — | |
| DittoModel Architecture=Unified (ResNet17)2024.06 | 67.4 | 56.17 | — | — | — | |
| LG-FedAvgModel Architecture=Unified (ResNet17)2024.06 | 64.6 | 54.76 | — | — | — | |
| RCLModel=ResNet-50, alpha=5, beta=202026.04 | 56 | 27.8 | 2.36 | -0.59 | -20 | |
| Focal LossModel=ViT-B162026.04 | 54.8 | 25.9 | 0.89 | — | — | |
| Focal LossModel=ResNet-502026.04 | 53.8 | 17.4 | 2.95 | — | — | |
| RCLModel=ViT-B16, alpha=5, beta=202026.04 | 47.8 | 10.1 | 0.59 | -0.3 | -33.3 |