mlsanityData to 14 Oct 2026

Adaptive Risk Minimization: Learning to Adapt to Domain Shift

arXiv 2007.02931 ↗First posted 6 Jul 2020cs.LG stat.MLcited by 183

Not on the map. Why?

Abstract (arXiv)

A fundamental assumption of most machine learning algorithms is that the training and test data are drawn from the same underlying distribution. However, this assumption is violated in almost all practical applications: machine learning systems are regularly tested under distribution shift, due to changing temporal correlations, atypical end users, or other factors. In this work, we consider the problem setting of domain generalization, where the training data are structured into domains and there may be multiple test time shifts, corresponding to new domains or domain distributions. Most prior methods aim to learn a single robust model or invariant feature space that performs well on all domains. In contrast, we aim to learn models that adapt at test time to domain shift using unlabeled test points. Our primary contribution is to introduce the framework of adaptive risk minimization (ARM), in which models are directly optimized for effective adaptation to shift by learning to adapt on the training domains. Compared to prior methods for robustness, invariance, and adaptation, ARM methods provide performance gains of 1-4% test accuracy on a number of image classification problems exhibiting domain shift.

Key facts. Cited by 183 papers in the map (data through 14 Oct 2026). First posted on arXiv 6 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 7; 2022: 26; 2023: 57; 2024: 43; 2025: 25; 2026: 25.

183 papers.

Sort:

By area:

  1. 2021MEMO: Test Time Robustness via Adaptation and Augmentationcited by 377
  2. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 359
  3. 2022Surgical Fine-Tuning Improves Adaptation to Distribution Shiftscited by 205
  4. 2024Efficient Test-Time Adaptation of Vision-Language Modelscited by 141
  5. 2022Domain Generalization by Mutual-Information Regularization with Pre-trained Modelscited by 133
  6. 2021The Norm Must Go On: Dynamic Unsupervised Domain Adaptation by Normalizationcited by 131
  7. 2022Sparse Mixture-of-Experts are Domain Generalizable Learnerscited by 85
  8. 2022Back to the Source: Diffusion-Driven Test-Time Adaptationcited by 71
  9. 2022Probable Domain Generalization via Quantile Risk Minimizationcited by 66
  10. 2023Advances and Challenges in Meta-Learning: A Technical Reviewcited by 65
  11. 2023On Pitfalls of Test-Time Adaptationcited by 64
  12. 2023Feed Two Birds with One Scone: Exploiting Wild Data for Both Out-of-Distribution Generalization and Detectioncited by 57
38 more
  1. 2022An Efficient Domain-Incremental Learning Approach to Drive in All Weather Conditionscited by 54
  2. 2022Meta-DMoE: Adapting to Domain Shift by Meta-Distillation from Mixture-of-Expertscited by 50
  3. 2023AdaNPC: Exploring Non-Parametric Classifier for Test-Time Adaptationcited by 48
  4. 2023Improved Test-Time Adaptation for Domain Generalizationcited by 47
  5. 2023MetaGCD: Learning to Continually Learn in Generalized Category Discoverycited by 42
  6. 2023Leveraging Vision-Language Models for Improving Domain Generalization in Image Classificationcited by 41
  7. 2022Self-Distilled Vision Transformer for Domain Generalizationcited by 35
  8. 2023Free Lunch for Domain Adversarial Training: Environment Label Smoothingcited by 33
  9. 2023Leveraging sparse and shared feature activations for disentangled representation learningcited by 32
  10. 2023Test-Time Domain Adaptation by Learning Domain-Aware Batch Normalizationcited by 31
  11. 2021Iterative Feature Matching: Toward Provable Domain Generalization with Logarithmic Environmentscited by 30
  12. 2021Meta-Calibration: Learning of Model Calibration Using Differentiable Expected Calibration Errorcited by 29
  13. 2023Domain Generalization in Computational Pathology: Survey and Guidelinescited by 29
  14. 2023Recurrence without Recurrence: Stable Video Landmark Detection with Deep Equilibrium Modelscited by 27
  15. 2023Decompose, Adjust, Compose: Effective Normalization by Playing with Frequency for Domain Generalizationcited by 26
  16. 2023Discovering environments with XRMcited by 26
  17. 2024FuRL: Visual-Language Models as Fuzzy Rewards for Reinforcement Learningcited by 26
  18. 2023Federated Domain Generalization: A Surveycited by 25
  19. 2023Learngene: Inheriting Condensed Knowledge from the Ancestry Model to Descendant Modelscited by 24
  20. 2024Rethinking Multi-domain Generalization with A General Learning Objectivecited by 23
  21. 2024Adapting to Distribution Shift by Visual Domain Prompt Generationcited by 23
  22. 2023MADG: Margin-based Adversarial Learning for Domain Generalizationcited by 22
  23. 2024DGMamba: Domain Generalization via Generalized State Space Modelcited by 21
  24. 2021Contrastive ACE: Domain Generalization Through Alignment of Causal Mechanismscited by 20
  25. 2022Causal Balancing for Domain Generalizationcited by 20
  26. 2023DART: Diversify-Aggregate-Repeat Training Improves Generalization of Neural Networkscited by 20
  27. 2023DomainAdaptor: A Novel Approach to Test-time Adaptationcited by 20
  28. 2023ERM++: An Improved Baseline for Domain Generalizationcited by 18
  29. 2023Benchmarking Test-Time Adaptation against Distribution Shifts in Image Classificationcited by 17
  30. 2023Rethinking Domain Generalization: Discriminability and Generalizabilitycited by 16
  31. 2024GM-DF: Generalized Multi-Scenario Deepfake Detectioncited by 16
  32. 2022Domain-Specific Risk Minimization for Out-of-Distribution Generalizationcited by 15
  33. 2023Energy-Based Test Sample Adaptation for Domain Generalizationcited by 14
  34. 2024Domain Generalisation via Imprecise Learningcited by 14
  35. 2025Learning to Adapt Frozen CLIP for Few-Shot Test-Time Domain Adaptationcited by 14
  36. 2023Multimodal Federated Learning in Healthcare: a Reviewcited by 13
  37. 2024HYPO: Hyperspherical Out-of-Distribution Generalizationcited by 13
  38. 2024Robust Fast Adaptation from Adversarially Explicit Task Distribution Generationcited by 13

50 most cited of 183 All 183 (JSON)