mlsanityData to 7 Oct 2026

WAT: Improve the Worst-class Robustness in Adversarial Training

arXiv 2302.04025 ↗First posted 8 Feb 2023cs.LGcited by 27

Not on the map. Why?

Abstract (arXiv)

Deep Neural Networks (DNN) have been shown to be vulnerable to adversarial examples. Adversarial training (AT) is a popular and effective strategy to defend against adversarial attacks. Recent works (Benz et al., 2020; Xu et al., 2021; Tian et al., 2021) have shown that a robust model well-trained by AT exhibits a remarkable robustness disparity among classes, and propose various methods to obtain consistent robust accuracy across classes. Unfortunately, these methods sacrifice a good deal of the average robust accuracy. Accordingly, this paper proposes a novel framework of worst-class adversarial training and leverages no-regret dynamics to solve this problem. Our goal is to obtain a classifier with great performance on worst-class and sacrifice just a little average robust accuracy at the same time. We then rigorously analyze the theoretical properties of our proposed algorithm, and the generalization error bound in terms of the worst-class robust risk. Furthermore, we propose a measurement to evaluate the proposed method in terms of both the average and worst-class accuracies. Experiments on various datasets and networks show that our proposed method outperforms the state-of-the-art approaches.

Key facts. Cited by 27 papers in the map (data through 7 Oct 2026). Builds on 20 of its 37 references that are in the map. First posted on arXiv 8 Feb 2023.

Builds on

20 of 37 references.

References · 20

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  4. 2013Intriguing properties of neural networkscited by 5,114
  5. 2016Wide Residual Networkscited by 3,057
  6. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  7. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  8. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  9. 2018Robustness May Be at Odds with Accuracycited by 873
  10. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  11. 2018Certified Defenses against Adversarial Examplescited by 305
  12. 2020Bag of Tricks for Adversarial Trainingcited by 199
8 more
  1. 2015Cost Sensitive Learning of Deep Feature Representations from Imbalanced Datacited by 167
  2. 2020To be Robust or to be Fair: Towards Fairness in Adversarial Trainingcited by 137
  3. 2018Rademacher Complexity for Adversarially Robust Generalizationcited by 132
  4. 2019VC Classes are Adversarially Robustly Learnable, but Only Improperlycited by 83
  5. 2020Do Wider Neural Networks Really Help Adversarial Robustness?cited by 79
  6. 2020Robustness May Be at Odds with Fairness: An Empirical Study on Class-wise Accuracycited by 45
  7. 2021Analysis and Applications of Class-wise Robustness in Adversarial Trainingcited by 33
  8. 2021CIFS: Improving Adversarial Robustness of CNNs via Channel-wise Importance-based Feature Selectioncited by 21

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 11; 2025: 8; 2026: 5.

27 papers.

Sort:

By area:

  1. 2023Better Diffusion Models Further Improve Adversarial Trainingcited by 207
  2. 2024DAFA: Distance-Aware Fair Adversarial Trainingcited by 11
  3. 2025Enhancing Robust Fairness via Confusional Spectral Regularizationcited by 11
  4. 2024Towards Fairness-Aware Adversarial Learningcited by 10
  5. 2023Improving Adversarial Robust Fairness via Anti-Bias Soft Label Distillationcited by 8
  6. 2024DRoP: Distributionally Robust Data Pruningcited by 7
  7. 2023Generalization Bounds for Adversarial Contrastive Learningcited by 6
  8. 2024Adversarial Training: A Surveycited by 6
  9. 2024Towards Adversarial Robustness via Debiased High-Confidence Logit Alignmentcited by 5
  10. 2024FAIR-TAT: Improving Model Fairness Using Targeted Adversarial Trainingcited by 5
  11. 2024Robustness Reprogramming for Representation Learningcited by 4
  12. 2024Exploring Frequencies via Feature Mixing and Meta-Learning for Improving Adversarial Transferabilitycited by 3
15 more
  1. 2026FragileFlow: Spectral Control of Correct-but-Fragile Predictions for Foundation Model Robustnesscited by 3
  2. 2024Improving Fast Adversarial Training via Self-Knowledge Guidancecited by 2
  3. 2024Combining Machine Learning Defenses without Conflictscited by 2
  4. 2025Towards Fair Class-wise Robustness: Class Optimal Distribution Adversarial Trainingcited by 2
  5. 2025Boosting Adversarial Robustness and Generalization with Dictionary Structurecited by 1
  6. 2026Taming the Long Tail: Rebalancing Adversarial Training via Adaptive Perturbationcited by 1
  7. 2024Improving Fast Adversarial Training Paradigm: An Example Taxonomy Perspectivecited by 0
  8. 2025Deep Minimax Classifiers for Imbalanced Datasets with a Small Number of Minority Samplescited by 0
  9. 2025Principal Eigenvalue Regularization for Improved Worst-Class Certified Robustness of Smoothed Classifierscited by 0
  10. 2025Towards Class-wise Fair Adversarial Training via Anti-Bias Soft Label Distillationcited by 0
  11. 2025TRIX- Trading Adversarial Fairness via Mixed Adversarial Trainingcited by 0
  12. 2025Sy-FAR: Symmetry-based Fair Adversarial Robustnesscited by 0
  13. 2026Reducing Class-Wise Performance Disparity via Margin Regularizationcited by 0
  14. 2026Adaptive Causal Alignment for High-Confidence Adversarial Trainingcited by 0
  15. 2026RL-FAT: Reinforcement Learning for Fair Adversarial Trainingcited by 0