mlsanityData to 7 Oct 2026

From Biased Selective Labels to Pseudo-Labels: An Expectation-Maximization Framework for Learning from Biased Decisions

arXiv 2406.18865 ↗First posted 27 Jun 2024cs.LG stat.MLcited by 4

Not on the map. Why?

Abstract (arXiv)

Selective labels occur when label observations are subject to a decision-making process; e.g., diagnoses that depend on the administration of laboratory tests. We study a clinically-inspired selective label problem called disparate censorship, where labeling biases vary across subgroups and unlabeled individuals are imputed as "negative" (i.e., no diagnostic test = no illness). Machine learning models naively trained on such labels could amplify labeling bias. Inspired by causal models of selective labels, we propose Disparate Censorship Expectation-Maximization (DCEM), an algorithm for learning in the presence of disparate censorship. We theoretically analyze how DCEM mitigates the effects of disparate censorship on model performance. We validate DCEM on synthetic data, showing that it improves bias mitigation (area between ROC curves) without sacrificing discriminative performance (AUC) compared to baselines. We achieve similar results in a sepsis classification task using clinical data.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 39 of its 100 references that are in the map. First posted on arXiv 27 Jun 2024.

Builds on

39 of 100 references.

References · 39

  1. 2018Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labelscited by 1,053
  2. 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 994
  3. 2017A Closer Look at Memorization in Deep Networkscited by 936
  4. 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 691
  5. 2016Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approachcited by 663
  6. 2019Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learningcited by 507
  7. 2021In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learningcited by 301
  8. 2019Adapting Neural Networks for the Estimation of Treatment Effectscited by 285
  9. 2019SELF: Learning to Filter Noisy Labels with Self-Ensemblingcited by 164
  10. 2019Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Ratescited by 142
  11. 2021Learning with Feature-Dependent Label Noise: A Progressive Approachcited by 102
  12. 2021Provably End-to-end Label-Noise Learning without Anchor Pointscited by 97
27 more
  1. 2020Beyond Class-Conditional Assumption: A Primary Attempt to Combat Instance-Dependent Label Noisecited by 82
  2. 2021Generalized Jensen-Shannon Divergence Loss for Learning with Noisy Labelscited by 74
  3. 2020Fair Classification with Group-Dependent Label Noisecited by 67
  4. 2021Clusterability as an Alternative to Anchor Points When Learning with Noisy Labelscited by 65
  5. 2019Recovering from Biased Data: Can Fairness Constraints Improve Accuracy?cited by 50
  6. 2021Instance-dependent Label-noise Learning under a Structural Causal Modelcited by 50
  7. 2018Beyond the Selected Completely At Random Assumption for Learning from Positive and Unlabeled Datacited by 36
  8. 2017Like trainer, like bot? Inheritance of bias in algorithmic content moderationcited by 34
  9. 2023Ground(less) Truth: A Causal Framework for Proxy Labels in Human-Algorithm Decision-Makingcited by 25
  10. 2020A Human-Centered Review of the Algorithms used within the U.S. Child Welfare Systemcited by 23
  11. 2019Causal Regularizationcited by 21
  12. 2022Instance-Dependent Noisy Label Learning via Graphical Modellingcited by 17
  13. 2022Centrality and Consistency: Two-Stage Clean Samples Identification for Learning with Instance-Dependent Noisy Labelscited by 16
  14. 2019What You See Is What You Get? The Impact of Representation Criteria on Human Bias in Hiringcited by 14
  15. 2020Does Fair Ranking Improve Minority Outcomes? Understanding the Interplay of Human and Algorithmic Biases in Online Hiringcited by 11
  16. 2022On Non-Random Missing Labels in Semi-Supervised Learningcited by 11
  17. 2023Counterfactual Prediction Under Outcome Measurement Errorcited by 10
  18. 2017Fast Threshold Tests for Detecting Discriminationcited by 9
  19. 2021Quantifying disparities in intimate partner violence: a machine learning method to correct for underreportingcited by 8
  20. 2023Domain constraints improve risk prediction when outcome data is missingcited by 8
  21. 2019Bias In, Bias Out? Evaluating the Folk Wisdomcited by 7
  22. 2022Disparate Censorship & Undertesting: A Source of Label Bias in Clinical Machine Learningcited by 7
  23. 2022Integrating Reward Maximization and Population Estimation: Sequential Decision-Making for Internal Revenue Service Audit Selectioncited by 5
  24. 2023Are labels informative in semi-supervised learning? -- Estimating and leveraging the missing-data mechanismcited by 5
  25. 2023Leveraging an Alignment Set in Tackling Instance-Dependent Label Noisecited by 4
  26. 2023Automatic Debiased Learning from Positive, Unlabeled, and Exposure Datacited by 1
  27. 2023Counterfactual Prediction Under Selective Confoundingcited by 1

Built on by

Cited by year (2026 to 7 Oct)

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

4 papers.

  1. 2026Beyond the Training Distribution: Evaluating Predictions Under Distribution Shift and Selection Biascited by 1
  2. 2025A Bayesian Model for Multi-stage Censoringcited by 0
  3. 2026Long-term Fairness with Selective Labelscited by 0
  4. 2026Causal Label Recovery in Payment Networkscited by 0