mlsanityData to 7 Oct 2026

Fair Bayes-Optimal Classifiers Under Predictive Parity

arXiv 2205.07182 ↗First posted 15 May 2022stat.ML cs.LGcited by 16

Not on the map. Why?

Abstract (arXiv)

Increasing concerns about disparate effects of AI have motivated a great deal of work on fair machine learning. Existing works mainly focus on independence- and separation-based measures (e.g., demographic parity, equality of opportunity, equalized odds), while sufficiency-based measures such as predictive parity are much less studied. This paper considers predictive parity, which requires equalizing the probability of success given a positive prediction among different protected groups. We prove that, if the overall performances of different groups vary only moderately, all fair Bayes-optimal classifiers under predictive parity are group-wise thresholding rules. Perhaps surprisingly, this may not hold if group performance levels vary widely; in this case we find that predictive parity among protected groups may lead to within-group unfairness. We then propose an algorithm we call FairBayes-DPP, aiming to ensure predictive parity when our condition is satisfied. FairBayes-DPP is an adaptive thresholding algorithm that aims to achieve predictive parity, while also seeking to maximize test accuracy. We provide supporting experiments conducted on synthetic and empirical data.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Builds on 36 of its 61 references that are in the map. First posted on arXiv 15 May 2022.

Builds on

36 of 61 references.

References · 36

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  5. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  6. 2014Going Deeper with Convolutionscited by 4,231
  7. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  8. 2014Sequence to Sequence Learning with Neural Networkscited by 2,974
  9. 2019XLNet: Generalized Autoregressive Pretraining for Language Understandingcited by 2,801
  10. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  11. 2014Certifying and removing disparate impactcited by 715
  12. 2016Fair prediction with disparate impact: A study of bias in recidivism prediction instrumentscited by 671
24 more
  1. 2018Mitigating Unwanted Biases with Adversarial Learningcited by 663
  2. 2017Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraintscited by 420
  3. 2017On Fairness and Calibrationcited by 406
  4. 2017Algorithmic decision making and the cost of fairnesscited by 402
  5. 2016Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatmentcited by 368
  6. 2018Learning Adversarially Fair and Transferable Representationscited by 353
  7. 2019Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigationcited by 252
  8. 2015The Variational Fair Autoencodercited by 223
  9. 2019Flexibly Fair Representation Learning by Disentanglementcited by 196
  10. 2016Fairness in Learning: Classic and Contextual Banditscited by 167
  11. 2018FairGAN: Fairness-aware Generative Adversarial Networkscited by 164
  12. 2020Fair Attribute Classification through Latent Space De-biasingcited by 125
  13. 2018Optimization with Non-Differentiable Constraints with Applications to Fairness, Recall, Churn, and Other Goalscited by 103
  14. 2018Achieving Fairness through Adversarial Learning: an Application to Recidivism Predictioncited by 77
  15. 2016A Confidence-Based Approach for Balancing Fairness and Accuracycited by 73
  16. 2020Learning Certified Individually Fair Representationscited by 69
  17. 2016Satisfying Real-world Goals with Dataset Constraintscited by 68
  18. 2018iFair: Learning Individually Fair Data Representations for Algorithmic Decision Makingcited by 64
  19. 2019Leveraging Labeled and Unlabeled Data for Consistent Fair Binary Classificationcited by 60
  20. 2016A statistical framework for fair predictive algorithmscited by 37
  21. 2021Classification with abstention but without disparitiescited by 27
  22. 2019Improved Adversarial Learning for Fair Classificationcited by 25
  23. 2022Bayes-Optimal Classifiers under Group Fairnesscited by 24
  24. 2020Fair Classification via Unconstrained Optimizationcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 5; 2024: 4; 2025: 3; 2026: 2.

16 papers.

Sort:

  1. 2022Bias Mitigation for Machine Learning Classifiers: A Comprehensive Surveycited by 125
  2. 2022Fair and Optimal Classification via Post-Processingcited by 50
  3. 2024Can LLMs Recognize Toxicity? A Structured Investigation Framework and Toxicity Metriccited by 23
  4. 2023Aleatoric and Epistemic Discrimination: Fundamental Limits of Fairness Interventionscited by 19
  5. 2023SYNAuG: Exploiting Synthetic Data for Data Imbalance Problemscited by 12
  6. 2023How Far Can Fairness Constraints Help Recover From Biased Data?cited by 5
  7. 2024Achieving Group Fairness through Independence in Predictive Process Monitoringcited by 5
  8. 2023Group-blind optimal transport to group parity and its constrained variantscited by 4
  9. 2024FairRR: Pre-Processing for Group Fairness through Randomized Responsecited by 2
  10. 2024Demographic parity in regression and classification within the unawareness frameworkcited by 2
  11. 2025On Optimal Steering to Achieve Exact Fairnesscited by 2
  12. 2023An Operational Perspective to Fairness Interventions: Where and How to Intervenecited by 1
4 more
  1. 2025Randomized multi-class classification under system constraints: a unified approach via post-processingcited by 1
  2. 2025EquiPy: Sequential Fairness using Optimal Transport in Pythoncited by 0
  3. 2026Fair Decisions from Calibrated Scores: Achieving Optimal Classification While Satisfying Sufficiencycited by 0
  4. 2026A Comprehensive View of Fairness through Distributional Stabilitycited by 0