mlsanityData to 7 Oct 2026

Beyond the Selected Completely At Random Assumption for Learning from Positive and Unlabeled Data

arXiv 1809.03207 ↗First posted 10 Sep 2018cs.LG stat.MLcited by 36

Not on the map. Why?

Abstract (arXiv)

Most positive and unlabeled data is subject to selection biases. The labeled examples can, for example, be selected from the positive set because they are easier to obtain or more obviously positive. This paper investigates how learning can be ena BHbled in this setting. We propose and theoretically analyze an empirical-risk-based method for incorporating the labeling mechanism. Additionally, we investigate under which assumptions learning is possible when the labeling mechanism is not fully understood and propose a practical method to enable this. Our empirical analysis supports the theoretical results and shows that taking into account the possibility of a selection bias, even when the labeling mechanism is unknown, improves the trained classifiers.

Key facts. Cited by 36 papers in the map (data through 7 Oct 2026). First posted on arXiv 10 Sep 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 4; 2023: 4; 2024: 11; 2025: 4; 2026: 8.

36 papers.

Sort:

By area:

  1. 2021Correcting Exposure Bias for Link Recommendationcited by 17
  2. 2022Doubly-Robust Estimation for Correcting Position-Bias in Click Feedback for Unbiased Learning to Rankcited by 17
  3. 2024Unbiased Learning to Rank Meets Reality: Lessons from Baidu's Large-Scale Search Datasetcited by 10
  4. 2023Mixture Proportion Estimation Beyond Irreducibilitycited by 9
  5. 2021Quantifying disparities in intimate partner violence: a machine learning method to correct for underreportingcited by 8
  6. 2024Semi-Supervised Disease Classification based on Limited Medical Image Datacited by 7
  7. 2025Learning from positive and unlabeled examples -Finite size sample boundscited by 6
  8. 2024From Lazy to Prolific: Tackling Missing Labels in Open Vocabulary Extreme Classification by Positive-Unlabeled Sequence Learningcited by 5
  9. 2024From Biased Selective Labels to Pseudo-Labels: An Expectation-Maximization Framework for Learning from Biased Decisionscited by 4
  10. 2022Double logistic regression approach to biased positive-unlabeled datacited by 3
  11. 2024Positive and Unlabeled Data: Model, Estimation, Inference, and Classificationcited by 3
  12. 2021Botcha: Detecting Malicious Non-Human Traffic in the Wildcited by 1
24 more
  1. 2021Stratified Learning: A General-Purpose Statistical Method for Improved Learning under Covariate Shiftcited by 1
  2. 2022Learning from Positive and Unlabeled Data with Augmented Classescited by 1
  3. 2023Automatic Debiased Learning from Positive, Unlabeled, and Exposure Datacited by 1
  4. 2023Single-sample versus case-control sampling scheme for Positive Unlabeled data: the story of two scenarioscited by 1
  5. 2024Verifying the Selected Completely at Random Assumption in Positive-Unlabeled Learningcited by 1
  6. 2024Novel Node Category Detection Under Subpopulation Shiftcited by 1
  7. 2025Unidentified and Confounded? Understanding Two-Tower Models for Unbiased Learning to Rankcited by 1
  8. 2025Accessible, Realistic, and Fair Evaluation of Positive-Unlabeled Learning Algorithmscited by 1
  9. 2026A proposal for PU classification under Non-SCAR using clustering and logistic modelcited by 1
  10. 2021Detecting corruption in single-bidder auctions via positive-unlabelled learningcited by 0
  11. 2022Risk bounds for PU learning under Selected At Random assumptioncited by 0
  12. 2023Positive Unlabeled Learning Selected Not At Random (PULSNAR): class proportion estimation when the SCAR assumption does not holdcited by 0
  13. 2024Soft Label PU Learningcited by 0
  14. 2024Augmented prediction of a true class for Positive Unlabeled data under selection biascited by 0
  15. 2024Learning from Concealed Labelscited by 0
  16. 2024ESA: Example Sieve Approach for Multi-Positive and Unlabeled Learningcited by 0
  17. 2025Open-Set Domain Adaptation Under Background Distribution Shift: Challenges and A Provably Efficient Solutioncited by 0
  18. 2026Focused PU learning from imbalanced datacited by 0
  19. 2026Quantifying and Auditing LLM Evaluation via Positive--Unlabeled Learningcited by 0
  20. 2026Dataset Usage Inference without Shadow Models or Held-out Datacited by 0
  21. 2026PUe: Biased Positive-Unlabeled Learning Enhancement by Causal Inferencecited by 0
  22. 2026AUC Maximization from Biased Positive-unlabeled Data with Confidencecited by 0
  23. 2026PU classification under Non-SCAR: clustering-assisted logistic model with oversampling enhancementcited by 0
  24. 2026GeoCFM: Positive-Only Conditional Flow Matching for Mineral Occurrence Samplingcited by 0