Adversarial Defense on CIFAR-10

Metric: Accuracy (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Hide extra data

Sort:

#	Model↕	Accuracy▼	Extra Data	Paper	Date↕	Code
1	WRN-28-10	90.03	Yes	Language Guided Adversarial Purification	2023-09-19	Code
2	Diffusion Classifier	89.85	No	Robust Classification via a Single Diffusion Model	2023-05-24	Code
3	Stochastic-LWTA/PGD/WideResNet-34-10	84.3	No	Stochastic Local Winner-Takes-All Networks Enabl...	2021-12-05	Code
4	Ours (Stochastic-LWTA/PGD/WideResNet-34-5)	83.4	No	Stochastic Local Winner-Takes-All Networks Enabl...	2021-12-05	Code
5	Ours (Stochastic-LWTA/PGD/WideResNet-34-1)	81.87	No	Stochastic Local Winner-Takes-All Networks Enabl...	2021-12-05	Code
6	ResNet18 (TRADES-ANCRA/PGD-40)	81.7	No	Enhancing Robust Representation in Adversarial T...	2023-10-05	Code
7	SLL X-Large	70.3	No	A Unified Algebraic Perspective on Lipschitz Neu...	2023-03-06	Code
8	SLL Large	69.8	No	A Unified Algebraic Perspective on Lipschitz Neu...	2023-03-06	Code
9	SLL Medium	69.1	No	A Unified Algebraic Perspective on Lipschitz Neu...	2023-03-06	Code
10	SLL Small	68.1	No	A Unified Algebraic Perspective on Lipschitz Neu...	2023-03-06	Code
11	PCL (against PGD, white box)	46.7	No	Adversarial Defense by Restricting the Hidden Sp...	2019-04-01	Code

#1WRN-28-10SOTA
90.03
Accuracy· Extra Data· 2023-09-19
Language Guided Adversarial Purification Code
#2Diffusion ClassifierSOTA
89.85
Accuracy· 2023-05-24
Robust Classification via a Single Diffusion Model Code
#3Stochastic-LWTA/PGD/WideResNet-34-10SOTA
84.3
Accuracy· 2021-12-05
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness Code
#4Ours (Stochastic-LWTA/PGD/WideResNet-34-5)
83.4
Accuracy· 2021-12-05
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness Code
#5Ours (Stochastic-LWTA/PGD/WideResNet-34-1)
81.87
Accuracy· 2021-12-05
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness Code
#6ResNet18 (TRADES-ANCRA/PGD-40)
81.7
Accuracy· 2023-10-05
Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria Code
#7SLL X-Large
70.3
Accuracy· 2023-03-06
A Unified Algebraic Perspective on Lipschitz Neural Networks Code
#8SLL Large
69.8
Accuracy· 2023-03-06
A Unified Algebraic Perspective on Lipschitz Neural Networks Code
#9SLL Medium
69.1
Accuracy· 2023-03-06
A Unified Algebraic Perspective on Lipschitz Neural Networks Code
#10SLL Small
68.1
Accuracy· 2023-03-06
A Unified Algebraic Perspective on Lipschitz Neural Networks Code
#11PCL (against PGD, white box)SOTA
46.7
Accuracy· 2019-04-01
Adversarial Defense by Restricting the Hidden Space of Deep Neural Networks Code