Image Classification on CIFAR-10 Raw
99.68Accuracydefensive patch generation framework
Evaluation results
| Method | Links | |
|---|---|---|
| defensive patch generation frameworkModel=MNet, Setting=Four models ensemble2022.04 | 99.68 | |
| defensive patch generation frameworkModel=VGG, Setting=Four models ensemble2022.04 | 99.27 | |
| defensive patch generation frameworkModel=SNet, Setting=Four models ensemble2022.04 | 99.02 | |
| defensive patch generation frameworkModel=RNet, Setting=Four models ensemble2022.04 | 98.82 | |
| UnAdvModel=RNet, Setting=Four models ensemble2022.04 | 95.51 | |
| VanillaModel=RNet, Setting=Four models ensemble2022.04 | 94.65 | |
| TransModel=RNet, Setting=Four models ensemble2022.04 | 93.69 | |
| VanillaModel=SNet, Setting=Four models ensemble2022.04 | 93.65 | |
| VanillaModel=VGG, Setting=Four models ensemble2022.04 | 92.67 | |
| VanillaModel=MNet, Setting=Four models ensemble2022.04 | 92.33 | |
| TransModel=MNet, Setting=Four models ensemble2022.04 | 91.31 | |
| TransModel=SNet, Setting=Four models ensemble2022.04 | 90.24 | |
| TransModel=VGG, Setting=Four models ensemble2022.04 | 88.84 | |
| UnAdvModel=VGG, Setting=Four models ensemble2022.04 | 88.57 | |
| UnAdvModel=SNet, Setting=Four models ensemble2022.04 | 82 | |
| UnAdvModel=MNet, Setting=Four models ensemble2022.04 | 73.14 |