Layer-wise Poisoning Attack (LPA) Robustness on Federated Learning Dataset
90.48AccuracyMulti-Krum
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Multi-KrumModel=CNN2026.02 | 90.48 | 70.84 | |
| FlameModel=CNN2026.02 | 90.37 | 65.57 | |
| FLTrustModel=CNN2026.02 | 90.16 | 84.5 | |
| FedAVGModel=CNN2026.02 | 89.29 | 83.45 | |
| MedianModel=VGG192026.02 | 78.63 | 93.15 | |
| MedianModel=ResNet182026.02 | 78.3 | 91.36 | |
| Trimmed-MeanModel=VGG192026.02 | 77.66 | 89.63 | |
| Trimmed-MeanModel=ResNet182026.02 | 76.18 | 90 | |
| FedAVGModel=VGG192026.02 | 75.9 | 89.71 | |
| FLTrustModel=VGG192026.02 | 74.01 | 68.05 | |
| FedAVGModel=ResNet182026.02 | 72.97 | 84.84 | |
| Multi-KrumModel=VGG192026.02 | 69.37 | 56.81 | |
| Multi-KrumModel=ResNet182026.02 | 68.45 | 84.16 | |
| FlameModel=ResNet182026.02 | 68.3 | 82.14 | |
| FlameModel=VGG192026.02 | 66.51 | 92.52 | |
| FLTrustModel=ResNet182026.02 | 64.95 | 83 |