Stochastic-LWTA/PGD/WideResNet-34-5

Reported on 4 benchmarks across 2 tasks · 1 paper

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

Adversarial4 results

Adversarial DefenseonCIFAR-10
Attack: AutoAttack· 2021-12-05
81.22
best: 82.6 (Stochastic-LWTA/PGD/WideResNet-34-10)
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness arXiv:2112.02671
Adversarial RobustnessonCIFAR-10
Accuracy· 2021-12-05
91.88
best: 95.23 (Mixed classifier)
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness arXiv:2112.02671
Adversarial RobustnessonCIFAR-10
Attack: AutoAttack· 2021-12-05
81.22
best: 82.6 (Stochastic-LWTA/PGD/WideResNet-34-10)
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness arXiv:2112.02671
Adversarial RobustnessonCIFAR-10
Robust Accuracy· 2021-12-05
83.4
best: 84.3 (Stochastic-LWTA/PGD/WideResNet-34-10)
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness arXiv:2112.02671