Personalized Federated Learning Unlearning on CIFAR-10 alpha_dir = 0.5
100PPSRetrain
Evaluation results
| Method | Links | ||
|---|---|---|---|
| RetrainpFL Architecture=FedPer2026.06 | 100 | 100 | |
| RetrainpFL Architecture=FedRep2026.06 | 100 | 100 | |
| RetrainpFL Architecture=Ditto2026.06 | 100 | 100 | |
| RetrainpFL Architecture=FedBN2026.06 | 100 | 100 | |
| pFedULpFL Architecture=Ditto2026.06 | 97.8 | 96.7 | |
| pFedULpFL Architecture=FedRep2026.06 | 97.5 | 96.3 | |
| pFedULpFL Architecture=FedPer2026.06 | 97.1 | 95.8 | |
| pFedULpFL Architecture=FedBN2026.06 | 96.8 | 95.1 | |
| Fine-tunepFL Architecture=Ditto2026.06 | 95.2 | 91.8 | |
| Fine-tunepFL Architecture=FedRep2026.06 | 94.6 | 91.1 | |
| Fine-tunepFL Architecture=FedPer2026.06 | 94.1 | 90.3 | |
| Fine-tunepFL Architecture=FedBN2026.06 | 93.4 | 89.5 | |
| FedEraser-adapt.pFL Architecture=Ditto2026.06 | 87.8 | 85.1 | |
| FedEraser-adapt.pFL Architecture=FedRep2026.06 | 86.9 | 83.9 | |
| FedEraser-adapt.pFL Architecture=FedPer2026.06 | 86.2 | 83.1 | |
| FedEraser-adapt.pFL Architecture=FedBN2026.06 | 85.3 | 82.2 | |
| GA-NaivepFL Architecture=Ditto2026.06 | 84.1 | 80.2 | |
| GA-NaivepFL Architecture=FedRep2026.06 | 83.3 | 79.1 | |
| GA-NaivepFL Architecture=FedPer2026.06 | 82.7 | 78.4 | |
| GA-NaivepFL Architecture=FedBN2026.06 | 81.9 | 77.1 |