mlsanityData to 7 Oct 2026

Self-Challenging Improves Cross-Domain Generalization

arXiv 2007.02454 ↗First posted 5 Jul 2020cs.CV cs.LGcited by 426

Not on the map. Why?

Abstract (arXiv)

Convolutional Neural Networks (CNN) conduct image classification by activating dominant features that correlated with labels. When the training and testing data are under similar distributions, their dominant features are similar, which usually facilitates decent performance on the testing data. The performance is nonetheless unmet when tested on samples from different distributions, leading to the challenges in cross-domain image classification. We introduce a simple training heuristic, Representation Self-Challenging (RSC), that significantly improves the generalization of CNN to the out-of-domain data. RSC iteratively challenges (discards) the dominant features activated on the training data, and forces the network to activate remaining features that correlates with labels. This process appears to activate feature representations applicable to out-of-domain data without prior knowledge of new domain and without learning extra network parameters. We present theoretical properties and conditions of RSC for improving cross-domain generalization. The experiments endorse the simple, effective and architecture-agnostic nature of our RSC method.

Key facts. Cited by 426 papers in the map (data through 7 Oct 2026). First posted on arXiv 5 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 74; 2022: 99; 2023: 107; 2024: 72; 2025: 56; 2026: 18.

426 papers.

Sort:

By area:

  1. 2021Generalizing to Unseen Domains: A Survey on Domain Generalizationcited by 762
  2. 2021Domain Generalization: A Surveycited by 760
  3. 2021SWAD: Domain Generalization by Seeking Flat Minimacited by 431
  4. 2021Towards Out-Of-Distribution Generalization: A Surveycited by 431
  5. 2021A Fourier-based Framework for Domain Generalizationcited by 287
  6. 2021FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Spacecited by 284
  7. 2021Gradient Matching for Domain Generalizationcited by 258
  8. 2021SelfReg: Self-supervised Contrastive Regularization for Domain Generalizationcited by 189
  9. 2021Fishr: Invariant Gradient Variances for Out-of-Distribution Generalizationcited by 186
  10. 2021Learning to Diversify for Single Domain Generalizationcited by 170
  11. 2021Deep Stable Learning for Out-Of-Distribution Generalizationcited by 153
  12. 2021Ensemble of Averages: Improving Model Selection and Boosting Performance in Domain Generalizationcited by 144
38 more
  1. 2022Causality Inspired Representation Learning for Domain Generalizationcited by 136
  2. 2022Domain Generalization by Mutual-Information Regularization with Pre-trained Modelscited by 133
  3. 2022Uncertainty Modeling for Out-of-Distribution Generalizationcited by 122
  4. 2021Causality-inspired Single-source Domain Generalization for Medical Image Segmentationcited by 118
  5. 2021Learning Debiased Representation via Disentangled Feature Augmentationcited by 116
  6. 2021Open Domain Generalization with Domain-Augmented Meta-Learningcited by 115
  7. 2021Model-Based Domain Generalizationcited by 103
  8. 2023Sharpness-Aware Gradient Matching for Domain Generalizationcited by 102
  9. 2021OoD-Bench: Quantifying and Understanding Two Dimensions of Out-of-Distribution Generalizationcited by 86
  10. 2021Invariant Information Bottleneck for Domain Generalizationcited by 86
  11. 2021Exploiting Domain-Specific Features to Enhance Domain Generalizationcited by 86
  12. 2022NICO++: Towards Better Benchmarking for Domain Generalizationcited by 86
  13. 2021Adversarially Adaptive Normalization for Single Domain Generalizationcited by 85
  14. 2022GLUE-X: Evaluating Natural Language Understanding Models from an Out-of-distribution Generalization Perspectivecited by 72
  15. 2023FedCLIP: Fast Generalization and Personalization for CLIP in Federated Learningcited by 70
  16. 2022Adversarial Style Augmentation for Domain Generalized Urban-Scene Segmentationcited by 69
  17. 2022Probable Domain Generalization via Quantile Risk Minimizationcited by 66
  18. 2023Attention-Challenging Multiple Instance Learning for Whole Slide Image Classificationcited by 61
  19. 2023PromptStyler: Prompt-driven Style Generation for Source-free Domain Generalizationcited by 59
  20. 2023Feed Two Birds with One Scone: Exploiting Wild Data for Both Out-of-Distribution Generalization and Detectioncited by 57
  21. 2022Prompt Vision Transformer for Domain Generalizationcited by 52
  22. 2022Learning Debiased Classifier with Biased Committeecited by 51
  23. 2021SAND-mask: An Enhanced Gradient Masking Strategy for the Discovery of Invariances in Domain Generalizationcited by 50
  24. 2021Test-time Batch Statistics Calibration for Covariate Shiftcited by 50
  25. 2022Pin the Memory: Learning to Generalize Semantic Segmentationcited by 48
  26. 2023AdaNPC: Exploring Non-Parametric Classifier for Test-Time Adaptationcited by 48
  27. 2023SimMMDG: A Simple and Effective Framework for Multi-modal Domain Generalizationcited by 48
  28. 2023Improved Test-Time Adaptation for Domain Generalizationcited by 45
  29. 2021Orthogonal Projection Losscited by 44
  30. 2022Rethinking Data Augmentation for Single-source Domain Generalization in Medical Image Segmentationcited by 44
  31. 2021Semi-Supervised Domain Generalization with Stochastic StyleMatchcited by 43
  32. 2023Meta-causal Learning for Single Domain Generalizationcited by 43
  33. 2021Federated Learning with Domain Generalizationcited by 41
  34. 2022FedDrive: Generalizing Federated Learning to Semantic Segmentation in Autonomous Drivingcited by 41
  35. 2022Deep Frequency Filtering for Domain Generalizationcited by 41
  36. 2023Leveraging Vision-Language Models for Improving Domain Generalization in Image Classificationcited by 41
  37. 2021Towards Unsupervised Domain Generalizationcited by 40
  38. 2021Feature Stylization and Domain-aware Contrastive Learning for Domain Generalizationcited by 40

50 most cited of 426 All 426 (JSON)