mlsanityData to 14 Oct 2026

Conflict-Aware Adversarial Training

arXiv 2410.16579 ↗First posted 21 Oct 2024cs.LG cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Adversarial training is the most effective method to obtain adversarial robustness for deep neural networks by directly involving adversarial samples in the training procedure. To obtain an accurate and robust model, the weighted-average method is applied to optimize standard loss and adversarial loss simultaneously. In this paper, we argue that the weighted-average method does not provide the best tradeoff for the standard performance and adversarial robustness. We argue that the failure of the weighted-average method is due to the conflict between the gradients derived from standard and adversarial loss, and further demonstrate such a conflict increases with attack budget theoretically and practically. To alleviate this problem, we propose a new trade-off paradigm for adversarial training with a conflict-aware factor for the convex combination of standard and adversarial loss, named Conflict-Aware Adversarial Training~(CA-AT). Comprehensive experimental results show that CA-AT consistently offers a superior trade-off between standard performance and adversarial robustness under the settings of adversarial training from scratch and parameter-efficient finetuning.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 47 of its 57 references that are in the map. First posted on arXiv 21 Oct 2024.

Builds on

47 of 57 references.

References · 47

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  5. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  6. 2022Training language models to follow instructions with human feedbackcited by 15,971
  7. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  8. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  9. 2017Mask R-CNNcited by 7,277
  10. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  11. 2016Wide Residual Networkscited by 3,070
  12. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
35 more
  1. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,517
  2. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,451
  3. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,318
  4. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,268
  5. 2020Gradient Surgery for Multi-Task Learningcited by 1,256
  6. 2018Efficient Lifelong Learning with A-GEMcited by 1,058
  7. 2016Practical Black-Box Attacks against Machine Learningcited by 903
  8. 2018Robustness May Be at Odds with Accuracycited by 880
  9. 2017Ensemble Adversarial Training: Attacks and Defensescited by 840
  10. 2020AdapterFusion: Non-Destructive Task Composition for Transfer Learningcited by 788
  11. 2019Big Transfer (BiT): General Visual Representation Learningcited by 769
  12. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 685
  13. 2016Delving into Transferable Adversarial Examples and Black-box Attackscited by 536
  14. 2021Conflict-Averse Gradient Descent for Multi-task Learningcited by 459
  15. 2020AdapterHub: A Framework for Adapting Transformerscited by 444
  16. 2019Minimally distorted Adversarial Examples with a Fast Adaptive Boundary Attackcited by 334
  17. 2021Recent Advances in Adversarial Training for Adversarial Robustnesscited by 279
  18. 2018Adversarial Attacks and Defences: A Surveycited by 259
  19. 2018Adversarial Logit Pairingcited by 224
  20. 2018Disentangling Adversarial Robustness and Generalizationcited by 172
  21. 2018MMA Training: Direct Input Space Margin Maximization through Adversarial Trainingcited by 172
  22. 2018Decoupling Direction and Norm for Efficient Gradient-Based L2 Adversarial Attacks and Defensescited by 109
  23. 2022LAS-AT: Adversarial Training with Learnable Attack Strategycited by 95
  24. 2018Curriculum Adversarial Trainingcited by 89
  25. 2023Revisiting Adversarial Training for ImageNet: Architectures, Training and Generalization across Threat Modelscited by 79
  26. 2020DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensemblescited by 73
  27. 2020CAT: Customized Adversarial Training for Improved Robustnesscited by 61
  28. 2021Domain Generalization via Gradient Surgerycited by 60
  29. 2019Instance adaptive adversarial training: Improved accuracy tradeoffs in neural netscited by 55
  30. 2022A Survey of Robust Adversarial Training in Pattern Recognition: Fundamental, Theory, and Methodologiescited by 44
  31. 2023Improving Fast Adversarial Training with Prior-Guided Knowledgecited by 35
  32. 2022Removing Batch Normalization Boosts Adversarial Trainingcited by 24
  33. 2023Initialization Matters for Adversarial Transfer Learningcited by 10
  34. 2023Restricted Orthogonal Gradient Projection for Continual Learningcited by 6
  35. 2022On Fragile Features and Batch Normalization in Adversarial Trainingcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Towards Universal Physical Adversarial Attacks via a Joint Multi-Objective and Multi-Model Optimization Frameworkcited by 0