Adversarial Robustness On Cifar 10
평가 지표
Accuracy
Attack: AutoAttack
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Accuracy | Attack: AutoAttack | Paper Title | Repository |
|---|---|---|---|---|
| GLOT-DR | 84.13 | 49.94 | Global-Local Regularization Via Distributional Robustness | |
| Mixed classifier | 95.23 | 68.06 | Improving the Accuracy-Robustness Trade-Off of Classifiers via Adaptive Smoothing | |
| Stochastic-LWTA/PGD/WideResNet-34-10 | 92.26 | 82.6 | Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness | |
| TRADES-ANCRA/ResNet18 | 81.70 | 59.70 | Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria | |
| Stochastic-LWTA/PGD/WideResNet-34-5 | 91.88 | 81.22 | Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness |
0 of 5 row(s) selected.