Layer-wise Selective Attack (LSA) Robustness on Federated Learning Dataset
90.58AccuracyMulti-Krum
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Multi-KrumModel=CNN2026.02 | 90.58 | 75.32 | |
| FlameModel=CNN2026.02 | 90.44 | 66.28 | |
| FLTrustModel=CNN2026.02 | 87.81 | 96.9 | |
| FedAVGModel=CNN2026.02 | 86.73 | 99.96 | |
| Multi-KrumModel=VGG192026.02 | 82.23 | 88.08 | |
| MedianModel=ResNet182026.02 | 79.61 | 93.25 | |
| Trimmed-MeanModel=ResNet182026.02 | 78.09 | 95.58 | |
| MedianModel=VGG192026.02 | 77.91 | 94.91 | |
| Multi-KrumModel=ResNet182026.02 | 76.89 | 92.01 | |
| Trimmed-MeanModel=VGG192026.02 | 76.63 | 94.01 | |
| FedAVGModel=VGG192026.02 | 76.56 | 96 | |
| FedAVGModel=ResNet182026.02 | 73.86 | 96.67 | |
| FLTrustModel=VGG192026.02 | 73.34 | 71.45 | |
| FlameModel=ResNet182026.02 | 72.19 | 96.98 | |
| FLTrustModel=ResNet182026.02 | 69.98 | 95.94 | |
| FlameModel=VGG192026.02 | 64.83 | 93 |