mlsanityData to 7 Oct 2026

Fair and Optimal Classification via Post-Processing

arXiv 2211.01528 ↗First posted 3 Nov 2022cs.LG cs.AI cs.CYcited by 50

Not on the map. Why?

Abstract (arXiv)

To mitigate the bias exhibited by machine learning models, fairness criteria can be integrated into the training process to ensure fair treatment across all demographics, but it often comes at the expense of model performance. Understanding such tradeoffs, therefore, underlies the design of fair algorithms. To this end, this paper provides a complete characterization of the inherent tradeoff of demographic parity on classification problems, under the most general multi-group, multi-class, and noisy setting. Specifically, we show that the minimum error rate achievable by randomized and attribute-aware fair classifiers is given by the optimal value of a Wasserstein-barycenter problem. On the practical side, our findings lead to a simple post-processing algorithm that derives fair classifiers from score functions, which yields the optimal fair classifier when the score is Bayes optimal. We provide suboptimality analysis and sample complexity for our algorithm, and demonstrate its effectiveness on benchmark datasets.

Key facts. Cited by 50 papers in the map (data through 7 Oct 2026). Builds on 40 of its 89 references that are in the map. First posted on arXiv 3 Nov 2022.

Builds on

40 of 89 references.

References · 40

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2017On Calibration of Modern Neural Networkscited by 4,860
  3. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  4. 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,304
  5. 2016Fair prediction with disparate impact: A study of bias in recidivism prediction instrumentscited by 671
  6. 2018Gender Bias in Coreference Resolution: Evaluation and Debiasing Methodscited by 657
  7. 2016CVXPY: A Python-Embedded Modeling Language for Convex Optimizationcited by 633
  8. 2016Inherent Trade-Offs in the Fair Determination of Risk Scorescited by 582
  9. 2021Retiring Adult: New Datasets for Fair Machine Learningcited by 443
  10. 2020Null It Out: Guarding Protected Attributes by Iterative Nullspace Projectioncited by 442
  11. 2017On Fairness and Calibrationcited by 406
  12. 2017Algorithmic decision making and the cost of fairnesscited by 402
28 more
  1. 2017Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairnesscited by 381
  2. 2021Persistent Anti-Muslim Bias in Large Language Modelscited by 376
  3. 2020Fairness in Machine Learning: A Surveycited by 369
  4. 2018Learning Adversarially Fair and Transferable Representationscited by 353
  5. 2019Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Settingcited by 346
  6. 2017Fairness in Criminal Justice Risk Assessments: The State of the Artcited by 315
  7. 2016Stochastic Optimization for Large-scale Optimal Transportcited by 188
  8. 2018Fair Clustering Through Fairletscited by 176
  9. 2019Fair Regression: Quantitative Definitions and Reduction-based Algorithmscited by 168
  10. 2016Fairness in Learning: Classic and Contextual Banditscited by 167
  11. 2019Inherent Tradeoffs in Learning Fair Representationscited by 144
  12. 2019Wasserstein Fair Classificationcited by 131
  13. 2017FA*IR: A Fair Top-k Ranking Algorithmcited by 113
  14. 2020Fair Regression with Wasserstein Barycenterscited by 99
  15. 2018Learning Controllable Fair Representationscited by 83
  16. 2019Scalable Fair Clusteringcited by 78
  17. 2019Conditional Learning of Fair Representationscited by 78
  18. 2016A Confidence-Based Approach for Balancing Fairness and Accuracycited by 73
  19. 2018Online Learning with an Unknown Fairness Metriccited by 61
  20. 2019Leveraging Labeled and Unlabeled Data for Consistent Fair Binary Classificationcited by 60
  21. 2019Average Individual Fairness: Algorithms, Generalization and Experimentscited by 46
  22. 2022Bayes-Optimal Classifiers under Group Fairnesscited by 24
  23. 2022Fair learning with Wasserstein barycenters for non-decomposable performance measurescited by 24
  24. 2016Fair Algorithms for Infinite and Contextual Banditscited by 16
  25. 2022Fair Bayes-Optimal Classifiers Under Predictive Paritycited by 16
  26. 2017Parallel Streaming Wasserstein Barycenterscited by 14
  27. 2019A gradual, semi-discrete approach to generative network training via explicit Wasserstein minimizationcited by 12
  28. 2021Towards Return Parity in Markov Decision Processescited by 12

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 7; 2024: 15; 2025: 17; 2026: 11.

50 papers.

Sort:

By area:

  1. 2023Revisiting Scalarization in Multi-Task Learning: A Theoretical Perspectivecited by 51
  2. 2024Representation Surgery: Theory and Practice of Affine Steeringcited by 47
  3. 2023FFB: A Fair Fairness Benchmark for In-Processing Group Fairness Methodscited by 33
  4. 2023FRAPPE: A Group Fairness Framework for Post-Processing Everythingcited by 16
  5. 2023Post-hoc Bias Scoring Is Optimal For Fair Classificationcited by 14
  6. 2024LibMOON: A Gradient-based MultiObjective OptimizatioN Library in PyTorchcited by 10
  7. 2024Counterfactual Fairness by Combining Factual and Counterfactual Predictionscited by 8
  8. 2024Differentially Private Post-Processing for Fair Regressioncited by 5
  9. 2024A Unifying Post-Processing Framework for Multi-Objective Learn-to-Defer Problemscited by 5
  10. 2024Post-Fair Federated Learning: Achieving Group and Community Fairness in Federated Learning via Post-processingcited by 4
  11. 2024Inducing Group Fairness in Prompt-Based Language Model Decisionscited by 3
  12. 2025Fairness-Aware Grouping for Continuous Sensitive Variables: Application for Debiasing Face Analysis with respect to Skin Tonecited by 3
38 more
  1. 2023Fair Supervised Learning with A Simple Random Sampler of Sensitive Attributescited by 2
  2. 2024Multi-Output Distributional Fairness via Post-Processingcited by 2
  3. 2024Demographic parity in regression and classification within the unawareness frameworkcited by 2
  4. 2024Efficient Fairness-Performance Pareto Front Computationcited by 2
  5. 2025The Cost of Local and Global Fairness in Federated Learningcited by 2
  6. 2025Enforcing Fair Predicted Scores on Intervals of Percentiles by Difference-of-Convex Constraintscited by 2
  7. 2025Meta Optimality for Demographic Parity Constrained Regression via Post-Processingcited by 2
  8. 2025FairAgent: Democratizing Fairness-Aware Machine Learning with LLM-Powered Agentscited by 2
  9. 2025Breaking the Dyadic Barrier: Rethinking Fairness in Link Prediction Beyond Demographic Paritycited by 2
  10. 2024UMOEA/D: A Multiobjective Evolutionary Algorithm for Uniform Pareto Objectives based on Decompositioncited by 1
  11. 2024OTTER: Effortless Label Distribution Adaptation of Zero-shot Modelscited by 1
  12. 2024AdapFair: Ensuring Adaptive Fairness for Machine Learning Operationscited by 1
  13. 2025Learning Pareto manifolds in high dimensions: How can regularization help?cited by 1
  14. 2025LoGoFair: Post-Processing for Local and Global Fairness in Federated Learningcited by 1
  15. 2025Bayes-Optimal Fair Classification with Multiple Sensitive Featurescited by 1
  16. 2025FairNet: Dynamic Fairness Correction without Performance Loss via Contrastive Conditional LoRAcited by 1
  17. 2025Randomized multi-class classification under system constraints: a unified approach via post-processingcited by 1
  18. 2026Optimal Fair Aggregation of Crowdsourced Noisy Labels using Demographic Parity Constraintscited by 1
  19. 2023An Empirical Study of Self-supervised Learning with Wasserstein Distancecited by 0
  20. 2023Constrained Reweighting of Distributions: an Optimal Transport Approachcited by 0
  21. 2024Fair Class-Incremental Learning using Sample Weightingcited by 0
  22. 2024Debiasify: Self-Distillation for Unsupervised Bias Mitigationcited by 0
  23. 2025A Critical Review of Predominant Bias in Neural Networkscited by 0
  24. 2025FedFACT: A Provable Framework for Controllable Group-Fairness Calibration in Federated Learningcited by 0
  25. 2025FairPOT: Balancing AUC Performance and Fairness with Proportional Optimal Transportcited by 0
  26. 2025FairAD: Computationally Efficient Fair Graph Clustering via Algebraic Distancecited by 0
  27. 2025Toward Unifying Group Fairness Evaluation from a Sparsity Perspectivecited by 0
  28. 2025Metanetworks as Regulatory Operators: Learning to Edit for Requirement Compliancecited by 0
  29. 2026Optimal Transport under Group Fairness Constraintscited by 0
  30. 2026Task-tailored Pre-processing: Fair Downstream Supervised Learningcited by 0
  31. 2026Operationalizing Fairness: Post-Hoc Threshold Optimization Under Hard Resource Limitscited by 0
  32. 2026Ethical Fairness in Ubiquitous Health Sensing without Known Attributescited by 0
  33. 2026A Model Ensemble-Based Post-Processing Framework for Fairness-Aware Predictioncited by 0
  34. 2026Demystifying the Optimal Fair Classifier in Multi-Class Classificationcited by 0
  35. 2026Fair Classification with Efficient and Post-hoc Controllable Fairness-Accuracy Trade-offcited by 0
  36. 2026Interpretable and Fair Generalized Additive Neural Networks via Multi-objective Learningcited by 0
  37. 2026A Comprehensive View of Fairness through Distributional Stabilitycited by 0
  38. 2026Sensitive-Topic Leakage Through LLM Routing Metadata: Measurement and Mitigationcited by 0