Image Classification on CIFAR-10 PUE (test)
90.09AccuracyOPA
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| OPABackbone=VGG19, Pre-trained=True (ImageNet)2024.06 | 90.09 | — | — | |
| Nonlinear transformation frameworkBackbone=VGG19, Pre-trained=True (ImageNet)2024.06 | 85.89 | -4.2 | -4.66 | |
| Adversarial trainingBackbone=VGG19, Pre-trained=True (ImageNet), Defense Method=PGD2024.06 | 75.35 | — | — | |
| No defenseBackbone=VGG19, Pre-trained=True (ImageNet)2024.06 | 11 | — | — |