Machine Unlearning on CIFAR-100 Forget 10%
75.67Test AccuracyBASE
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| BASEBackbone=ResNet-182026.03 | 75.67 | 73.97 | 12.24 | |
| SSDBackbone=ResNet-182026.03 | 75.67 | 73.97 | 12.24 | |
| NEGGRADBackbone=ResNet-182026.03 | 74.55 | 74.35 | 12.11 | |
| RETRAINBackbone=ResNet-182026.03 | 74.36 | 50.71 | 0 | |
| NEGGRAD+Backbone=ResNet-182026.03 | 74.23 | 73.72 | 11.74 | |
| SALUNBackbone=ResNet-182026.03 | 67.74 | 55.26 | 6.71 | |
| REGUNBackbone=ResNet-182026.03 | 66.55 | 50.1 | 4.8 | |
| AMUNBackbone=ResNet-182026.03 | 65.73 | 50.75 | 7.11 | |
| FINETUNEBackbone=ResNet-182026.03 | 65.57 | 61.33 | 6.96 | |
| ℓ1-SPARSEBackbone=ResNet-182026.03 | 65.55 | 61.02 | 6.83 |