Machine Unlearning on CIFAR-100 (forget)
0.2457KLoM (Mean)NegGrad (SAM, SAM)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| NegGrad (SAM, SAM)Pre-training Optimizer (A)=SAM, Unlearning Optimizer (U)=SAM, Forget set difficulty (F)=mid, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.2457 | 0.8111 | |
| NegGrad (SAM, SAM)Pre-training Optimizer (A)=SAM, Unlearning Optimizer (U)=SAM, Forget set difficulty (F)=high, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.1714 | 0.4901 | |
| NegGrad (SGD, SAM)Pre-training Optimizer (A)=SGD, Unlearning Optimizer (U)=SAM, Forget set difficulty (F)=mid, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.1589 | 0.5681 | |
| NegGrad (SAM, SGD)Pre-training Optimizer (A)=SAM, Unlearning Optimizer (U)=SGD, Forget set difficulty (F)=high, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.1549 | 0.4901 | |
| NegGrad (SAM, SGD)Pre-training Optimizer (A)=SAM, Unlearning Optimizer (U)=SGD, Forget set difficulty (F)=mid, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.1513 | 0.5371 | |
| NegGrad (SGD, SAM)Pre-training Optimizer (A)=SGD, Unlearning Optimizer (U)=SAM, Forget set difficulty (F)=high, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.1384 | 0.3397 | |
| NegGrad (SGD, SGD)Pre-training Optimizer (A)=SGD, Unlearning Optimizer (U)=SGD, Forget set difficulty (F)=high, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.1294 | 0.3397 | |
| NegGrad (SGD, SGD)Pre-training Optimizer (A)=SGD, Unlearning Optimizer (U)=SGD, Forget set difficulty (F)=mid, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.1221 | 0.4959 | |
| NegGrad (SGD, SGD)Pre-training Optimizer (A)=SGD, Unlearning Optimizer (U)=SGD, Forget set difficulty (F)=low, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.0284 | 0.0956 | |
| NegGrad (SAM, SGD)Pre-training Optimizer (A)=SAM, Unlearning Optimizer (U)=SGD, Forget set difficulty (F)=low, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.025 | 0.0956 | |
| NegGrad (SGD, SAM)Pre-training Optimizer (A)=SGD, Unlearning Optimizer (U)=SAM, Forget set difficulty (F)=low, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.0163 | 0.0956 | |
| NegGrad (SAM, SAM)Pre-training Optimizer (A)=SAM, Unlearning Optimizer (U)=SAM, Forget set difficulty (F)=low, Backbone=ResNet50, Unlearning algorithm=NegGrad2025.06 | 0.015 | 0.0956 |