Adversarial Defense

Competitions with currently unpublished results:

TrojAI

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 26–50 of 403 papers

Title	Date	Tasks	Status	Hype
Threat Model-Agnostic Adversarial Defense using Diffusion Models	Jul 17, 2022	Adversarial DefenseDenoising	CodeCode Available	1
Perturbation Inactivation Based Adversarial Defense for Face Recognition	Jul 13, 2022	Adversarial AttackAdversarial Defense	CodeCode Available	1
CARLA-GeAR: a Dataset Generator for a Systematic Evaluation of Adversarial Robustness of Vision Models	Jun 9, 2022	Adversarial DefenseAdversarial Robustness	CodeCode Available	1
Self-recoverable Adversarial Examples: A New Effective Protection Mechanism in Social Networks	Apr 26, 2022	Adversarial AttackAdversarial Defense	CodeCode Available	1
GUARD: Graph Universal Adversarial Defense	Apr 20, 2022	Adversarial Defense	CodeCode Available	1
CgAT: Center-Guided Adversarial Training for Deep Hashing-Based Retrieval	Apr 18, 2022	Adversarial AttackAdversarial Defense	CodeCode Available	1
LPF-Defense: 3D Adversarial Defense based on Frequency Analysis	Feb 23, 2022	3D Point Cloud ClassificationAdversarial Defense	CodeCode Available	1
Open-set Adversarial Defense with Clean-Adversarial Mutual Learning	Feb 12, 2022	Adversarial DefenseDenoising	CodeCode Available	1
Adversarial Attack and Defense of YOLO Detectors in Autonomous Driving Scenarios	Feb 10, 2022	Adversarial AttackAdversarial Defense	CodeCode Available	1
Layer-wise Regularized Adversarial Training using Layers Sustainability Analysis (LSA) framework	Feb 5, 2022	Adversarial AttackAdversarial Defense	CodeCode Available	1
Revisiting and Advancing Fast Adversarial Training Through The Lens of Bi-Level Optimization	Dec 23, 2021	Adversarial Defense	CodeCode Available	1
Segment and Complete: Defending Object Detectors against Adversarial Patch Attacks with Robust Patch Detection	Dec 8, 2021	Adversarial Attack DetectionAdversarial Defense	CodeCode Available	1
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness	Dec 5, 2021	Adversarial AttackAdversarial Defense	CodeCode Available	1
Person Re-identification Method Based on Color Attack and Joint Defence	Nov 18, 2021	Adversarial DefenseMetric Learning	CodeCode Available	1
DropAttack: A Masked Weight Adversarial Training Method to Improve Generalization of Neural Networks	Aug 29, 2021	Adversarial AttackAdversarial Defense	CodeCode Available	1
RAILS: A Robust Adversarial Immune-inspired Learning System	Jun 27, 2021	Adversarial DefenseAdversarial Robustness	CodeCode Available	1
Helper-based Adversarial Training: Reducing Excessive Margin to Achieve a Better Accuracy vs. Robustness Trade-off	Jun 18, 2021	Adversarial DefenseAdversarial Robustness	CodeCode Available	1
The art of defense: letting networks fool the attacker	Apr 7, 2021	Adversarial Defense	CodeCode Available	1
Fast Certified Robust Training with Short Warmup	Mar 31, 2021	Adversarial Defense	CodeCode Available	1
LiBRe: A Practical Bayesian Approach to Adversarial Detection	Mar 27, 2021	Adversarial DefenseUncertainty Quantification	CodeCode Available	1
Sandwich Batch Normalization: A Drop-In Replacement for Feature Distribution Heterogeneity	Feb 22, 2021	Adversarial DefenseConditional Image Generation	CodeCode Available	1
Eliminate Deviation with Deviation for Data Augmentation and a General Multi-modal Data Learning Method	Jan 21, 2021	Adversarial DefenseData Augmentation	CodeCode Available	1
A Person Re-identification Data Augmentation Method with Adversarial Defense Effect	Jan 21, 2021	Adversarial DefenseData Augmentation	CodeCode Available	1
Towards Adversarial Robustness of Bayesian Neural Network through Hierarchical Variational Inference	Jan 1, 2021	Adversarial DefenseAdversarial Robustness	CodeCode Available	1
Geometric Adversarial Attacks and Defenses on 3D Point Clouds	Dec 10, 2020	Adversarial AttackAdversarial Defense	CodeCode Available	1

Show:10 25 50

← PrevPage 2 of 17Next →

All datasets CIFAR-10 ImageNet (non-targeted PGD, max perturbation=4)ImageNet ImageNet (targeted PGD, max perturbation=16)MNIST CAAD 2018 miniImageNet

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	WRN-28-10	Accuracy	90.03	—	Unverified
2	Diffusion Classifier	Accuracy	89.85	—	Unverified
3	Stochastic-LWTA/PGD/WideResNet-34-10	Accuracy	84.3	—	Unverified
4	Ours (Stochastic-LWTA/PGD/WideResNet-34-5)	Accuracy	83.4	—	Unverified
5	Ours (Stochastic-LWTA/PGD/WideResNet-34-1)	Accuracy	81.87	—	Unverified
6	ResNet18 (TRADES-ANCRA/PGD-40)	Accuracy	81.7	—	Unverified
7	Stochastic-LWTA/PGD/WideResNet-34-5	Attack: AutoAttack	81.22	—	Unverified
8	PCL (against PGD, white box)	Accuracy	46.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SAT-EfficientNet-L1	Accuracy	58.6	—	Unverified
2	LLR-ResNet-152	Accuracy	47	—	Unverified
3	ResNet-152 free-m=4	Accuracy	36	—	Unverified
4	ResNet-101 free-m=4	Accuracy	34.3	—	Unverified
5	ResNet-50 free-m=4	Accuracy	31.8	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ResNet101	Accuracy	99.8	—	Unverified
2	InceptionV3	Accuracy	98.6	—	Unverified
3	Feature Denoising	Accuracy	49.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ResNet-152 Denoise	Accuracy	42.8	—	Unverified
2	ResNeXt-101 DenoiseAll	Accuracy	40.4	—	Unverified
3	ResNet-152	Accuracy	39	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Defense GAN	Accuracy	0.85	—	Unverified
2	PuVAE	Accuracy	0.81	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Feature Denoising	Accuracy	50.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Auto Encoder-Block Switching defense with GradCAM	Accuracy	88.54	—	Unverified