Label Unlearning on MedMNIST PathMNIST (test)
84.78Test AccuracyCVFU
Evaluation results
| Method | Links | |
|---|---|---|
| CVFUm (number of unlearning classes)=2, n (percentage of samples to unlearn)=100%2025.12 | 84.78 | |
| Retrainm (number of unlearning classes)=2, n (percentage of samples to unlearn)=100%2025.12 | 84.54 | |
| ICOm (number of unlearning classes)=2, n (percentage of samples to unlearn)=100%2025.12 | 83.25 | |
| GAm (number of unlearning classes)=2, n (percentage of samples to unlearn)=100%2025.12 | 81.89 | |
| Retrainm (number of unlearning classes)=5, n (percentage of samples to unlearn)=100%2025.12 | 81.75 | |
| FedORAm (number of unlearning classes)=2, n (percentage of samples to unlearn)=100%2025.12 | 80.25 | |
| FedORAm (number of unlearning classes)=5, n (percentage of samples to unlearn)=100%2025.12 | 78.33 | |
| ICOm (number of unlearning classes)=5, n (percentage of samples to unlearn)=100%2025.12 | 77.44 | |
| GAm (number of unlearning classes)=5, n (percentage of samples to unlearn)=100%2025.12 | 74.15 | |
| CVFUm (number of unlearning classes)=5, n (percentage of samples to unlearn)=100%2025.12 | 73.3 |