mlsanityData to 7 Oct 2026

Online Adaptation to Label Distribution Shift

arXiv 2107.04520 ↗First posted 9 Jul 2021cs.LG stat.MLcited by 44

Not on the map. Why?

Abstract (arXiv)

Machine learning models often encounter distribution shifts when deployed in the real world. In this paper, we focus on adaptation to label distribution shift in the online setting, where the test-time label distribution is continually changing and the model must dynamically adapt to it without observing the true label. Leveraging a novel analysis, we show that the lack of true label does not hinder estimation of the expected test loss, which enables the reduction of online label shift adaptation to conventional online learning. Informed by this observation, we propose adaptation algorithms inspired by classical online learning techniques such as Follow The Leader (FTL) and Online Gradient Descent (OGD) and derive their regret bounds. We empirically verify our findings under both simulated and real world label distribution shifts and show that OGD is particularly effective and robust to a variety of challenging label shift scenarios.

Key facts. Cited by 44 papers in the map (data through 7 Oct 2026). Builds on 15 of its 45 references that are in the map. First posted on arXiv 9 Jul 2021.

Builds on

15 of 45 references.

References · 15

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2017On Calibration of Modern Neural Networkscited by 4,860
  3. 2016Concrete Problems in AI Safetycited by 1,582
  4. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,239
  5. 2019Learning Imbalanced Datasets with Label-Distribution-Aware Margin Losscited by 987
  6. 2019Contrastive Adaptation Network for Unsupervised Domain Adaptationcited by 407
  7. 2015Return of Frustratingly Easy Domain Adaptationcited by 389
  8. 2018Detecting and Correcting for Label Shift with Black Box Predictorscited by 364
  9. 2018DeepJDOT: Deep Joint Distribution Optimal Transport for Unsupervised Domain Adaptationcited by 212
  10. 2020Domain Adaptation with Conditional Distribution Matching and Generalized Label Shiftcited by 124
  11. 2020A Unified View of Label Shift Estimationcited by 119
  12. 2019Regularized Learning for Domain Adaptation under Label Shiftscited by 118
3 more
  1. 2016Beyond Sharing Weights for Deep Domain Adaptationcited by 80
  2. 2018Online Model Distillation for Efficient Video Inferencecited by 60
  3. 2018Algorithms and Theory for Multiple-Source Adaptationcited by 53

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 7; 2023: 13; 2024: 10; 2025: 9; 2026: 4.

44 papers.

Sort:

By area:

  1. 2021ReAct: Out-of-distribution Detection With Rectified Activationscited by 486
  2. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 353
  3. 2023Towards Real-World Test-Time Adaptation: Tri-Net Self-Training with Balanced Normalizationcited by 34
  4. 2023Can You Rely on Your Model Evaluation? Improving Model Evaluation with Synthetic Test Datacited by 31
  5. 2023Active learning for data streams: a surveycited by 29
  6. 2022Adapting to Online Label Shift with Provable Guaranteescited by 26
  7. 2024Any-Shift Prompting for Generalization over Distributionscited by 24
  8. 2023Adaptive Test-Time Personalization for Federated Learningcited by 23
  9. 2022Test-time Adaptation with Calibration of Medical Image Classification Nets for Label Distribution Shiftcited by 20
  10. 2022Estimating and Explaining Model Performance When Both Covariates and Labels Shiftcited by 19
  11. 2023Adapting to Continuous Covariate Shift via Online Density Ratio Estimationcited by 15
  12. 2026On-the-Fly VLA Adaptation via Test-Time Reinforcement Learningcited by 14
32 more
  1. 2022Video Unsupervised Domain Adaptation with Deep Learning: A Comprehensive Surveycited by 12
  2. 2022Adapting to Latent Subgroup Shifts via Concepts and Proxiescited by 12
  3. 2023Online Label Shift: Optimal Dynamic Regret meets Practical Algorithmscited by 11
  4. 2023Capturing Conversion Rate Fluctuation during Sales Promotions: A Novel Historical Data Reuse Approachcited by 8
  5. 2024Fully Test-time Adaptation for Tabular Datacited by 8
  6. 2024AdapTable: Test-Time Adaptation for Tabular Data via Shift-Aware Uncertainty Calibrator and Label Distribution Handlercited by 6
  7. 2022Time-Varying Propensity Score to Bridge the Gap between the Past and Presentcited by 5
  8. 2023Prominent Roles of Conditionally Invariant Components in Domain Adaptation: Theory and Algorithmscited by 5
  9. 2025When Shift Happens - Confounding Is to Blamecited by 5
  10. 2022A Multi-stage Framework with Mean Subspace Computation and Recursive Feedback for Online Unsupervised Domain Adaptationcited by 3
  11. 2025ASAP: Unsupervised Post-training with Label Distribution Shift Adaptive Learning Ratecited by 3
  12. 2023Learning to Adapt to Online Streams with Distribution Shiftscited by 2
  13. 2024Familiarity: Better Evaluation of Zero-Shot Named Entity Recognition by Quantifying Label Shifts in Synthetic Training Datacited by 2
  14. 2025Adapting to Online Distribution Shifts in Deep Learning: A Black-Box Approachcited by 2
  15. 2025OASIS: Open-world Adaptive Self-supervised and Imbalanced-aware Systemcited by 2
  16. 2026Fed-ADE: Adaptive Learning Rate for Federated Post-adaptation under Distribution Shiftcited by 2
  17. 2023Detecting and Learning Out-of-Distribution Data in the Open world: Algorithm and Theorycited by 1
  18. 2024Online Feature Updates Improve Online (Generalized) Label Shift Adaptationcited by 1
  19. 2024A Comprehensive Review of Machine Learning Advances on Data Change: A Cross-Field Perspectivecited by 1
  20. 2025Open Set Label Shift with Test Time Out-of-Distribution Referencecited by 1
  21. 2025Knowledge Distillation of Uncertainty using Deep Latent Factor Modelcited by 1
  22. 2025Bayesian-based Online Label Shift Estimation with Dynamic Dirichlet Priorscited by 1
  23. 2023A Dataset Fusion Algorithm for Generalised Anomaly Detection in Homogeneous Periodic Time Series Datasetscited by 0
  24. 2023Prediction under Latent Subgroup Shifts with High-Dimensional Observationscited by 0
  25. 2024Fairness Hub Technical Briefs: Definition and Detection of Distribution Shiftcited by 0
  26. 2024Online Resource Allocation for Edge Intelligence with Colocated Model Retraining and Inferencecited by 0
  27. 2024An adaptive transfer learning perspective on classification in non-stationary environmentscited by 0
  28. 2024Theory-inspired Label Shift Adaptation via Aligned Distribution Mixturecited by 0
  29. 2025Machine Unlearning for Streaming Forgettingcited by 0
  30. 2025Should Bias be Eliminated? A General Framework to Use Bias for OOD Generalizationcited by 0
  31. 2026FMMO: Detecting the Divergence Between Local Attribution and Global Driftcited by 0
  32. 2026From matrix inversion to constraints: provably tighter confidence regions for importance weights in label shiftcited by 0