mlsanityData to 14 Oct 2026

The Statistical Fairness-Accuracy Frontier

arXiv 2508.17622 ↗First posted 25 Aug 2025stat.ML cs.LG econ.THcited by 3

Not on the map. Why?

Abstract (arXiv)

We study fairness-accuracy tradeoffs when a single predictive model must serve multiple demographic groups. A useful tool for understanding this tradeoff is the fairness-accuracy (FA) Pareto frontier, which characterizes the set of models that cannot be improved in either fairness or accuracy without worsening the other. While characterizing the FA frontier requires full knowledge of the data distribution, we focus on the finite-sample regime, quantifying how well a designer can approximate any point on the frontier from limited data and bounding the worst-case gap. In particular, we derive worst-case-optimal estimators that depend on the designer's knowledge of the covariate distribution. For each estimator, we characterize how finite-sample effects asymmetrically impact each group's welfare and identify optimal sample allocation strategies. Finally, we provide uniform finite-sample bounds for the entire FA frontier, yielding confidence bands that quantify the reliability of welfare comparisons across alternative fairness-accuracy tradeoffs.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 34 of its 62 references that are in the map. First posted on arXiv 25 Aug 2025.

Builds on

34 of 62 references.

References · 34

  1. 2013Distributed Representations of Words and Phrases and their Compositionalitycited by 3,474
  2. 2016Equality of Opportunity in Supervised Learningcited by 2,007
  3. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,245
  4. 2018A Reductions Approach to Fair Classificationcited by 587
  5. 2016Inherent Trade-Offs in the Fair Determination of Risk Scorescited by 586
  6. 2017Algorithmic decision making and the cost of fairnesscited by 403
  7. 2018Empirical Risk Minimization under Fairness Constraintscited by 226
  8. 2019Mitigating Bias in Algorithmic Hiring: Evaluating Claims and Practicescited by 174
  9. 2019Fair Regression: Quantitative Definitions and Reduction-based Algorithmscited by 168
  10. 2020Minimax Pareto Fairness: A Multi Objective Perspectivecited by 153
  11. 2017A Convex Framework for Fair Regressioncited by 148
  12. 2020Fair Regression with Wasserstein Barycenterscited by 100
22 more
  1. 2016A Confidence-Based Approach for Balancing Fairness and Accuracycited by 73
  2. 2019Recovering from Biased Data: Can Fairness Constraints Improve Accuracy?cited by 50
  3. 2019Fair Classification and Social Welfarecited by 40
  4. 2019Group Retention when Using Machine Learning in Sequential Decision Making: the Interplay between User Dynamics and Fairnesscited by 38
  5. 2019Exact minimax risk for linear least squares, and the lower tail of sample covariance matricescited by 34
  6. 2021Emergent Unfairness in Algorithmic Fairness-Accuracy Trade-Off Researchcited by 34
  7. 2020The Fairness-Accuracy Pareto Frontcited by 26
  8. 2019General Fair Empirical Risk Minimizationcited by 23
  9. 2021On the Properties of Kullback-Leibler Divergence Between Multivariate Gaussian Distributionscited by 23
  10. 2023Aleatoric and Epistemic Discrimination: Fundamental Limits of Fairness Interventionscited by 19
  11. 2020Balancing Competing Objectives with Noisy Data: Score-Based Classifiers for Welfare-Aware Machine Learningcited by 17
  12. 2024Utility-Fairness Trade-Offs and How to Find Themcited by 15
  13. 2018A General Framework for Fair Regressioncited by 13
  14. 2021Learning to be Fair: A Consequentialist Approach to Equitable Decision-Makingcited by 11
  15. 2022To the Fairness Frontier and Beyond: Identifying, Quantifying, and Optimizing the Fairness-Accuracy Pareto Frontiercited by 11
  16. 2023Loss Balancing for Fair Supervised Learningcited by 11
  17. 2021Costs and Benefits of Fair Regressioncited by 8
  18. 2024What Constitutes a Less Discriminatory Algorithm?cited by 8
  19. 2024Inference for an Algorithmic Fairness-Accuracy Frontiercited by 2
  20. 2025Welfare as a Guiding Principle for Machine Learning -- From Compass, to Lens, to Roadmapcited by 2
  21. 2024Algorithmic Fairness and Social Welfarecited by 1
  22. 2025Selecting for Less Discriminatory Algorithms: A Relational Search Framework for Navigating Fairness-Accuracy Trade-offs in Practicecited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025Statistical Guarantees in the Search for Less Discriminatory Algorithmscited by 0
  2. 2026Fair Regression under Demographic Parity: A Unified Frameworkcited by 0
  3. 2026Identification and Inference for Algorithmic Frontiers with Selective Labelscited by 0