Image Classification on CIFAR-10 (test) (Adversarial Robustness - Natural/FGSM)
68.52Accuracy (Natural)Federated AT with Data Sharing
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Federated AT with Data SharingNumber of clients (K)=5, Communication rounds (R)=150, Local epochs (E)=1, Dirichlet (beta)=0.1, Attack type=White-box2025.12 | 68.52 | 54.66 | |
| CalFATNumber of clients (K)=5, Communication rounds (R)=150, Local epochs (E)=1, Dirichlet (beta)=0.1, Attack type=White-box2025.12 | 64.69 | 35.02 |