ResNet18 (TRADES-ANCRA/PGD-40)

Reported on 3 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Adversarial3 results

Adversarial DefenseonCIFAR-10
Robust Accuracy· 2023-10-05
82.96
SOTA
Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria arXiv:2310.03358
Adversarial DefenseonCIFAR-10
Accuracy· 2023-10-05
81.7
best: 90.03 (WRN-28-10)
Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria arXiv:2310.03358
Adversarial DefenseonCIFAR-10
Attack: AutoAttack· 2023-10-05
59.7
best: 82.6 (Stochastic-LWTA/PGD/WideResNet-34-10)
Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria arXiv:2310.03358