mlsanityData to 7 Oct 2026

Unsupervised Robust Domain Adaptation without Source Data

arXiv 2103.14577 ↗First posted 26 Mar 2021cs.CVcited by 19

Not on the map. Why?

Abstract (arXiv)

We study the problem of robust domain adaptation in the context of unavailable target labels and source data. The considered robustness is against adversarial perturbations. This paper aims at answering the question of finding the right strategy to make the target model robust and accurate in the setting of unsupervised domain adaptation without source data. The major findings of this paper are: (i) robust source models can be transferred robustly to the target; (ii) robust domain adaptation can greatly benefit from non-robust pseudo-labels and the pair-wise contrastive loss. The proposed method of using non-robust pseudo-labels performs surprisingly well on both clean and adversarial samples, for the task of image classification. We show a consistent performance improvement of over 10% in accuracy against the tested baselines on four benchmark datasets.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). Builds on 39 of its 56 references that are in the map. First posted on arXiv 26 Mar 2021.

Builds on

39 of 56 references.

References · 39

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  7. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  8. 2020Supervised Contrastive Learningcited by 2,803
  9. 2014Unsupervised Domain Adaptation by Backpropagationcited by 1,906
  10. 2016Adversarial examples in the physical worldcited by 1,762
  11. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  12. 2018Moment Matching for Multi-Source Domain Adaptationcited by 1,327
27 more
  1. 2015Learning Transferable Features with Deep Adaptation Networkscited by 1,323
  2. 2017Adversarial Discriminative Domain Adaptationcited by 1,309
  3. 2017Deep Hashing Network for Unsupervised Domain Adaptationcited by 1,181
  4. 2017Deeper, Broader and Artier Domain Generalizationcited by 983
  5. 2020Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptationcited by 844
  6. 2015The Limitations of Deep Learning in Adversarial Settingscited by 784
  7. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
  8. 2017ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Modelscited by 656
  9. 2016Deep Transfer Learning with Joint Adaptation Networkscited by 609
  10. 2017Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Modelscited by 459
  11. 2018Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Modelscited by 362
  12. 2018A Survey of Unsupervised Deep Domain Adaptationcited by 361
  13. 2017Defense against Adversarial Attacks Using High-Level Representation Guided Denoisercited by 329
  14. 2020Do Adversarially Robust ImageNet Models Transfer Better?cited by 315
  15. 2020Universal Source-Free Domain Adaptationcited by 219
  16. 2018Towards the first adversarially robust neural network model on MNISTcited by 104
  17. 2021Domain Impression: A Source Data Free Domain Adaptation Methodcited by 91
  18. 2020Domain Adaptation without Source Datacited by 88
  19. 2019Adversarially robust transfer learningcited by 66
  20. 2017Adversarial Transformation Networks: Learning to Generate Adversarial Examplescited by 60
  21. 2019Adversarial Defense by Restricting the Hidden Space of Deep Neural Networkscited by 56
  22. 2017Defense against Universal Adversarial Perturbationscited by 48
  23. 2018Towards Imperceptible and Robust Adversarial Example Attacks against Neural Networkscited by 38
  24. 2017Biologically inspired protection of deep networks from adversarial attackscited by 23
  25. 2020Adversarially-Trained Deep Nets Transfer Better: Illustration on Image Classificationcited by 18
  26. 2020Unsupervised Domain Adaptation in the Absence of Source Datacited by 14
  27. 2019Texture Underfitting for Domain Adaptationcited by 1

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 5; 2023: 7; 2024: 4; 2025: 2.

19 papers.

Sort:

  1. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 353
  2. 2023A Comprehensive Survey on Source-free Domain Adaptationcited by 104
  3. 2022Source-Free Unsupervised Domain Adaptation: A Surveycited by 79
  4. 2022Improving Test-Time Adaptation via Shift-agnostic Weight Regularization and Nearest Source Prototypescited by 67
  5. 2022Source-free Depth for Object Pop-outcited by 52
  6. 2021Data-Free Knowledge Transfer: A Surveycited by 31
  7. 2023TeSLA: Test-Time Self-Learning With Automatic Adversarial Augmentationcited by 28
  8. 2023Adversarial Alignment for Source Free Object Detectioncited by 25
  9. 2022Cross-Modal Knowledge Transfer Without Task-Relevant Source Datacited by 17
  10. 2022SALAD: Source-free Active Label-Agnostic Domain Adaptation for Classification, Segmentation and Detectioncited by 10
  11. 2024Source-Free Domain Adaptation of Weakly-Supervised Object Localization Models for Histologycited by 9
  12. 2023Source-Free Domain Adaptation for RGB-D Semantic Segmentation with Vision Transformerscited by 8
7 more
  1. 2025Source-Free Object Detection with Detection Transformercited by 4
  2. 2023AdaptGuard: Defending Against Universal Attacks for Model Adaptationcited by 3
  3. 2023Open-World Pose Transfer via Sequential Test-Time Adaptioncited by 3
  4. 2024Key Design Choices in Source-Free Unsupervised Domain Adaptation: An In-depth Empirical Analysiscited by 1
  5. 2025Grad-CL: Source Free Domain Adaptation with Gradient Guided Feature Disalignmentcited by 1
  6. 2024Trust And Balance: Few Trusted Samples Pseudo-Labeling and Temperature Scaled Loss for Effective Source-Free Unsupervised Domain Adaptationcited by 0
  7. 2024LEARN: A Unified Framework for Multi-Task Domain Adapt Few-Shot Learningcited by 0