Federated Learning Classification on CIFAR-100 non-iid Dirichlet 0.6 (test)
55.84AccuracyFedSpeed+MAN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| FedSpeed+MANRegularizer=MAN2025.12 | 55.84 | 153 | |
| FedDyn+MANRegularizer=MAN2025.12 | 55.63 | 143 | |
| FedDC+MANRegularizer=MAN2025.12 | 55.4 | 129 | |
| FedDCRegularizer=None2025.12 | 52.64 | 304 | |
| FedSAM+MANRegularizer=MAN2025.12 | 52.62 | 281 | |
| FedAvg+MANRegularizer=MAN2025.12 | 52.42 | 210 | |
| FedASAM+MANRegularizer=MAN2025.12 | 51.89 | 351 | |
| FedSpeedRegularizer=None2025.12 | 51.33 | 390 | |
| FedDynRegularizer=None2025.12 | 49.91 | 486 | |
| FedASAMRegularizer=None2025.12 | 45.48 | 500 | |
| FedSAMRegularizer=None2025.12 | 43.36 | 500 | |
| FedAvgRegularizer=None2025.12 | 40.39 | 500 |