mlsanityData to 14 Oct 2026

Robustness to Subpopulation Shift with Domain Label Noise via Regularized Annotation of Domains

arXiv 2402.11039 ↗First posted 16 Feb 2024cs.LG stat.MLcited by 7

Not on the map. Why?

Abstract (arXiv)

Existing methods for last layer retraining that aim to optimize worst-group accuracy (WGA) rely heavily on well-annotated groups in the training data. We show, both in theory and practice, that annotation-based data augmentations using either downsampling or upweighting for WGA are susceptible to domain annotation noise, and in high-noise regimes approach the WGA of a model trained with vanilla empirical risk minimization. We introduce Regularized Annotation of Domains (RAD) in order to train robust last layer classifiers without the need for explicit domain annotations. Our results show that RAD is competitive with other recently proposed domain annotation-free techniques. Most importantly, RAD outperforms state-of-the-art annotation-reliant methods even with only 5% noise in the training data for several publicly available datasets.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 19 of its 24 references that are in the map. First posted on arXiv 16 Feb 2024.

Builds on

19 of 24 references.

References · 19

  1. 2014Deep Learning Face Attributes in the Wildcited by 3,693
  2. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,682
  3. 2019Invariant Risk Minimizationcited by 1,854
  4. 2016Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approachcited by 665
  5. 2021Just Train Twice: Improving Group Robustness without Training Group Informationcited by 523
  6. 2019Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classificationcited by 404
  7. 2022Last Layer Re-Training is Sufficient for Robustness to Spurious Correlationscited by 381
  8. 2019Label Propagation for Deep Semi-supervised Learningcited by 289
  9. 2021Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotationscited by 256
  10. 2022Improving Out-of-Distribution Robustness via Selective Augmentationcited by 202
  11. 2021Simple data balancing achieves competitive worst-group-accuracycited by 178
  12. 2022On Feature Learning in the Presence of Spurious Correlationscited by 145
7 more
  1. 2019Distributionally Robust Language Modelingcited by 143
  2. 2023Change is Hard: A Closer Look at Subpopulation Shiftcited by 120
  3. 2023Simple and Fast Group Robustness by Automatic Feature Reweightingcited by 68
  4. 2022Learning with Neighbor Consistency for Noisy Labelscited by 55
  5. 2023Towards Last-layer Retraining for Group Robustness with Fewer Annotationscited by 54
  6. 2022Why does Throwing Away Data Improve Worst-Group Error?cited by 22
  7. 2022Improving group robustness under noisy labels using predictive uncertaintycited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 1; 2026: 3.

7 papers.

  1. 2024Towards Formalizing Spuriousness of Biased Datasets Using Partial Information Decompositioncited by 6
  2. 2024The Group Robustness is in the Details: Revisiting Finetuning under Spurious Correlationscited by 6
  3. 2025On the Unreasonable Effectiveness of Last-layer Retrainingcited by 3
  4. 2024Label Noise Robustness for Domain-Agnostic Fair Corrections via Nearest Neighbors Label Spreadingcited by 0
  5. 2026ProME: Prototype-Margin Environments with Repair-Aware Selection for Group-Robust Learningcited by 0
  6. 2026Optimal Transport Reweighting for Robust Learning under Spurious Correlations and Label Noisecited by 0
  7. 2026Distributionally Robust Survival Models under Subpopulation Shift and Outlier Contaminationcited by 0