mlsanityData to 7 Oct 2026

To be Robust or to be Fair: Towards Fairness in Adversarial Training

arXiv 2010.06121 ↗First posted 13 Oct 2020cs.LG stat.MLcited by 137

Not on the map. Why?

Abstract (arXiv)

Adversarial training algorithms have been proved to be reliable to improve machine learning models' robustness against adversarial examples. However, we find that adversarial training algorithms tend to introduce severe disparity of accuracy and robustness between different groups of data. For instance, a PGD adversarially trained ResNet18 model on CIFAR-10 has 93% clean accuracy and 67% PGD l-infty-8 robust accuracy on the class "automobile" but only 65% and 17% on the class "cat". This phenomenon happens in balanced datasets and does not exist in naturally trained models when only using clean samples. In this work, we empirically and theoretically show that this phenomenon can happen under general adversarial training algorithms which minimize DNN models' robust errors. Motivated by these findings, we propose a Fair-Robust-Learning (FRL) framework to mitigate this unfairness problem when doing adversarial defenses. Experimental results validate the effectiveness of FRL.

Key facts. Cited by 137 papers in the map (data through 7 Oct 2026). First posted on arXiv 13 Oct 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 11; 2022: 29; 2023: 36; 2024: 30; 2025: 23; 2026: 8.

137 papers.

Sort:

By area:

  1. 2021Trustworthy AI: From Principles to Practicescited by 206
  2. 2021Data Collection and Quality Challenges in Deep Learning: A Data-Centric AI Perspectivecited by 116
  3. 2022A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learningcited by 104
  4. 2021Trustworthy AI: A Computational Perspectivecited by 93
  5. 2021Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Trainingcited by 58
  6. 2023CFA: Class-wise Calibrated Fair Adversarial Trainingcited by 53
  7. 2022Diagnosing failures of fairness transfer across distribution shift in real-world medical settingscited by 45
  8. 2022Pruning has a disparate impact on model accuracycited by 43
  9. 2022A Comprehensive Survey on Trustworthy Recommender Systemscited by 32
  10. 2023WAT: Improve the Worst-class Robustness in Adversarial Trainingcited by 27
  11. 2022Fast Adversarial Training with Adaptive Step Sizecited by 24
  12. 2024On the Duality Between Sharpness-Aware Minimization and Adversarial Trainingcited by 24
38 more
  1. 2024Deconstructing The Ethics of Large Language Models from Long-standing Issues to New-emerging Dilemmas: A Surveycited by 23
  2. 2021Algorithm Fairness in AI for Medicine and Healthcarecited by 21
  3. 2024The Better Angels of Machine Personality: How Personality Relates to LLM Safetycited by 21
  4. 2022CalFAT: Calibrated Federated Adversarial Training with Label Skewnesscited by 20
  5. 2023Mitigating Test-Time Bias for Fair Image Retrievalcited by 20
  6. 2021The Authors Matter: Understanding and Mitigating Implicit Bias in Deep Text Classificationcited by 18
  7. 2024Towards Tracing Trustworthiness Dynamics: Revisiting Pre-training Period of Large Language Modelscited by 18
  8. 2022Improving Robust Fairness via Balance Adversarial Trainingcited by 17
  9. 2023Unlocking Accuracy and Fairness in Differentially Private Image Classificationcited by 16
  10. 2024LLM Whisperer: An Inconspicuous Attack to Bias LLM Responsescited by 15
  11. 2023Sharpness-Aware Minimization Alone can Improve Adversarial Robustnesscited by 14
  12. 2022A Review of Speech-centric Trustworthy Machine Learning: Privacy, Safety, and Fairnesscited by 13
  13. 2024Revisiting Adversarial Training under Long-Tailed Distributionscited by 13
  14. 2024The Pursuit of Fairness in Artificial Intelligence Models: A Surveycited by 13
  15. 2022Can Adversarial Training Be Manipulated By Non-Robust Features?cited by 12
  16. 2023Robustness Implies Fairness in Causal Algorithmic Recoursecited by 12
  17. 2023A Classification of Feedback Loops and Their Relation to Biases in Automated Decision-Making Systemscited by 12
  18. 2021Imbalanced Adversarial Training with Reweightingcited by 11
  19. 2022Improving Adversarial Robustness by Putting More Regularizations on Less Robust Samplescited by 11
  20. 2023Unlocking the Potential of ChatGPT: A Comprehensive Exploration of its Applications, Advantages, Limitations, and Future Directions in Natural Language Processingcited by 11
  21. 2024DAFA: Distance-Aware Fair Adversarial Trainingcited by 11
  22. 2025Enhancing Robust Fairness via Confusional Spectral Regularizationcited by 11
  23. 2022Towards A Holistic View of Bias in Machine Learning: Bridging Algorithmic Fairness and Imbalanced Learningcited by 10
  24. 2022Improving Privacy-Preserving Vertical Federated Learning by Efficient Communication with ADMMcited by 10
  25. 2024Towards Fairness-Aware Adversarial Learningcited by 10
  26. 2025Robustness and Cybersecurity in the EU Artificial Intelligence Actcited by 10
  27. 2021Fairness Properties of Face Recognition and Obfuscation Systemscited by 9
  28. 2021Unfairness Despite Awareness: Group-Fair Classification with Strategic Agentscited by 9
  29. 2022Understanding the Impact of Adversarial Robustness on Accuracy Disparitycited by 9
  30. 2023A Comprehensive Study on Dataset Distillation: Performance, Privacy, Robustness and Fairnesscited by 9
  31. 2024Demonstration of an Adversarial Attack Against a Multimodal Vision Language Model for Pathology Imagingcited by 9
  32. 2024AIM: Attributing, Interpreting, Mitigating Data Unfairnesscited by 9
  33. 2021NoiLIn: Improving Adversarial Training and Correcting Stereotype of Noisy Labelscited by 8
  34. 2022A.I. Robustness: a Human-Centered Perspective on Technological Challenges and Opportunitiescited by 8
  35. 2023Muffin: A Framework Toward Multi-Dimension AI Fairness by Uniting Off-the-Shelf Modelscited by 8
  36. 2023Improving Adversarial Robust Fairness via Anti-Bias Soft Label Distillationcited by 8
  37. 2024The Pitfalls and Promise of Conformal Inference Under Adversarial Attackscited by 8
  38. 2023Toward Operationalizing Pipeline-aware ML Fairness: A Research Agenda for Developing Practical Guidelines and Toolscited by 7

50 most cited of 137 All 137 (JSON)