Image Classification on CIFAR-10 (Retain)
100AccuracyRetrain
Evaluation results
| Method | Links | ||
|---|---|---|---|
| RetrainBackbone=ResNet-182025.03 | 100 | 0 | |
| l1-sparseBackbone=ResNet-182025.03 | 99.98 | 0.02 | |
| BadTBackbone=ResNet-182025.03 | 99.89 | 0.11 | |
| NoTBackbone=ResNet-182025.03 | 99.69 | 0.31 | |
| SSDBackbone=ResNet-182025.03 | 98.82 | 1.18 | |
| GABackbone=ResNet-182025.03 | 98.76 | 1.24 | |
| FTBackbone=ResNet-182025.03 | 98.63 | 1.37 | |
| SalUnBackbone=ResNet-182025.03 | 98.29 | 1.71 | |
| RandLBackbone=ResNet-182025.03 | 94.34 | 5.66 |