Image Classification on CIFAR-10 AdvP
70.81Accuracydefensive patch generation framework
Evaluation results
| Method | Links | |
|---|---|---|
| defensive patch generation frameworkModel=RNet, Setting=Four models ensemble2022.04 | 70.81 | |
| defensive patch generation frameworkModel=SNet, Setting=Four models ensemble2022.04 | 67.67 | |
| defensive patch generation frameworkModel=MNet, Setting=Four models ensemble2022.04 | 64.82 | |
| TransModel=RNet, Setting=Four models ensemble2022.04 | 59.52 | |
| VanillaModel=RNet, Setting=Four models ensemble2022.04 | 53.51 | |
| TransModel=MNet, Setting=Four models ensemble2022.04 | 53.17 | |
| TransModel=SNet, Setting=Four models ensemble2022.04 | 52.91 | |
| UnAdvModel=RNet, Setting=Four models ensemble2022.04 | 49.39 | |
| VanillaModel=MNet, Setting=Four models ensemble2022.04 | 46.76 | |
| VanillaModel=SNet, Setting=Four models ensemble2022.04 | 43.76 | |
| defensive patch generation frameworkModel=VGG, Setting=Four models ensemble2022.04 | 40.83 | |
| TransModel=VGG, Setting=Four models ensemble2022.04 | 33.18 | |
| UnAdvModel=SNet, Setting=Four models ensemble2022.04 | 31.71 | |
| UnAdvModel=VGG, Setting=Four models ensemble2022.04 | 29.24 | |
| VanillaModel=VGG, Setting=Four models ensemble2022.04 | 19.99 | |
| UnAdvModel=MNet, Setting=Four models ensemble2022.04 | 19.8 |