mlsanityData to 7 Oct 2026

GAT: Guided Adversarial Training with Pareto-optimal Auxiliary Tasks

arXiv 2302.02907 ↗First posted 6 Feb 2023cs.CV cs.CR cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

While leveraging additional training data is well established to improve adversarial robustness, it incurs the unavoidable cost of data collection and the heavy computation to train models. To mitigate the costs, we propose Guided Adversarial Training (GAT), a novel adversarial training technique that exploits auxiliary tasks under a limited set of training data. Our approach extends single-task models into multi-task models during the min-max optimization of adversarial training, and drives the loss optimization with a regularization of the gradient curvature across multiple tasks. GAT leverages two types of auxiliary tasks: self-supervised tasks, where the labels are generated automatically, and domain-knowledge tasks, where human experts provide additional labels. Experimentally, GAT increases the robust AUC of CheXpert medical imaging dataset from 50% to 83% and On CIFAR-10, GAT outperforms eight state-of-the-art adversarial training and achieves 56.21% robust accuracy with Resnet-50. Overall, we demonstrate that guided multi-task learning is an actionable and promising avenue to push further the boundaries of model robustness.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 40 of its 68 references that are in the map. First posted on arXiv 6 Feb 2023.

Builds on

40 of 68 references.

References · 40

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2016Wide Residual Networkscited by 3,057
  6. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  7. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  8. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,586
  9. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  10. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  11. 2019CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparisoncited by 1,410
  12. 2016Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzlescited by 1,379
28 more
  1. 2020Gradient Surgery for Multi-Task Learningcited by 1,210
  2. 2016Adversarial Machine Learning at Scalecited by 934
  3. 2018Multi-Task Learning as Multi-Objective Optimizationcited by 809
  4. 2017An Overview of Multi-Task Learning in Deep Neural Networkscited by 791
  5. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  6. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 682
  7. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 659
  8. 2019Adversarial Training for Free!cited by 638
  9. 2019Using Self-Supervised Learning Can Improve Model Robustness and Uncertaintycited by 511
  10. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  11. 2019Which Tasks Should Be Learned Together in Multi-task Learning?cited by 367
  12. 2017Certifying Some Distributional Robustness with Principled Adversarial Trainingcited by 333
  13. 2019Minimally distorted Adversarial Examples with a Fast Adaptive Boundary Attackcited by 333
  14. 2018Adversarially Robust Generalization Requires More Datacited by 318
  15. 2019Searching for A Robust Neural Architecture in Four GPU Hourscited by 258
  16. 2020Attacks Which Do Not Kill Training Make Adversarial Learning Strongercited by 242
  17. 2021Improving Robustness using Generated Datacited by 233
  18. 2021Data Augmentation Can Improve Robustnesscited by 193
  19. 2018Robustness via curvature regularization, and vice versacited by 174
  20. 2020Gradient Vaccine: Investigating and Improving Multi-task Optimization in Massively Multilingual Modelscited by 167
  21. 2020Torchattacks: A PyTorch Repository for Adversarial Attackscited by 156
  22. 2020Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuningcited by 152
  23. 2019Adversarial Sensor Attack on LiDAR-based Perception in Autonomous Drivingcited by 151
  24. 2019Understanding Adversarial Attacks on Deep Learning Based Medical Image Analysis Systemscited by 151
  25. 2020On the limits of cross-domain generalization in automated X-ray predictioncited by 61
  26. 2020MaxUp: A Simple Way to Improve Generalization of Neural Network Trainingcited by 27
  27. 2020Improved Noise and Attack Robustness for Semantic Segmentation by Using Multi-Task Training with Self-Supervised Depth Estimationcited by 26
  28. 2021Adversarial Robustness in Multi-Task Learning: Promises and Illusionscited by 17

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 2.

4 papers.

  1. 2024SafePowerGraph: Safety-aware Evaluation of Graph Neural Networks for Transmission Power Gridscited by 5
  2. 2023Attacking All Tasks at Once Using Adversarial Examples in Multi-Task Learningcited by 1
  3. 2023Hazards in Deep Learning Testing: Prevalence, Impact and Recommendationscited by 0
  4. 2024Enhancing Adversarial Robustness through Multi-Objective Representation Learningcited by 0