mlsanityData to 14 Oct 2026

Fairness Overfitting in Machine Learning: An Information-Theoretic Perspective

arXiv 2506.07861 ↗First posted 9 Jun 2025cs.LG cs.AI cs.ITcited by 3

Not on the map. Why?

Abstract (arXiv)

Despite substantial progress in promoting fairness in high-stake applications using machine learning models, existing methods often modify the training process, such as through regularizers or other interventions, but lack formal guarantees that fairness achieved during training will generalize to unseen data. Although overfitting with respect to prediction performance has been extensively studied, overfitting in terms of fairness loss has received far less attention. This paper proposes a theoretical framework for analyzing fairness generalization error through an information-theoretic lens. Our novel bounding technique is based on Efron-Stein inequality, which allows us to derive tight information-theoretic fairness generalization bounds with both Mutual Information (MI) and Conditional Mutual Information (CMI). Our empirical results validate the tightness and practical relevance of these bounds across diverse fairness-aware learning algorithms. Our framework offers valuable insights to guide the design of algorithms improving fairness generalization.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 43 of its 79 references that are in the map. First posted on arXiv 9 Jun 2025.

Builds on

43 of 79 references.

References · 43

  1. 2016Equality of Opportunity in Supervised Learningcited by 2,007
  2. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,917
  3. 2016Fair prediction with disparate impact: A study of bias in recidivism prediction instrumentscited by 672
  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. 2018Learning Adversarially Fair and Transferable Representationscited by 353
  8. 2017Information-theoretic analysis of generalization capability of learning algorithmscited by 303
  9. 2020Reasoning About Generalization via Conditional Mutual Informationcited by 147
  10. 2019Wasserstein Fair Classificationcited by 131
  11. 2019Tightening Mutual Information Based Bounds on Generalization Errorcited by 117
  12. 2021Fair Mixup: Fairness via Interpolationcited by 109
31 more
  1. 2021Information-Theoretic Generalization Bounds for Stochastic Gradient Descentcited by 84
  2. 2020Robust Fairness under Covariate Shiftcited by 58
  3. 2021Information-theoretic generalization bounds for black-box learning algorithmscited by 49
  4. 2019Average Individual Fairness: Algorithms, Generalization and Experimentscited by 46
  5. 2021Controllable Guarantees for Fair Outcomes via Contrastive Information Estimationcited by 45
  6. 2021Generalization Bounds For Meta-Learning: An Information-Theoretic Analysiscited by 45
  7. 2021Tighter expected generalization error bounds via Wasserstein distancecited by 43
  8. 2022Achieving Fairness at No Utility Cost via Data Reweighing with Influencecited by 38
  9. 2019Transfer of Machine Learning Fairness across Domainscited by 36
  10. 2017Estimating Mutual Information for Discrete-Continuous Mixturescited by 34
  11. 2021Selective Regression Under Fairness Criteriacited by 34
  12. 2023FFB: A Fair Fairness Benchmark for In-Processing Group Fairness Methodscited by 33
  13. 2020Information-Theoretic Generalization Bounds for Meta-Learning and Applicationscited by 32
  14. 2022On Learning Fairness and Accuracy on Multiple Subgroupscited by 32
  15. 2021On the Generalization of Models Trained with SGD: Information-Theoretic Bounds and Implicationscited by 31
  16. 2020Individually Conditional Individual Mutual Information Bound on Generalization Errorcited by 30
  17. 2022Fairness Transferability Subject to Bounded Distribution Shiftcited by 28
  18. 2022A New Family of Generalization Bounds Using Samplewise Evaluated CMIcited by 28
  19. 2023Fairness and Accuracy under Domain Generalizationcited by 24
  20. 2020Information-theoretic analysis for transfer learningcited by 23
  21. 2023Tighter Information-Theoretic Generalization Bounds from Supersamplescited by 22
  22. 2022Fair Ranking with Noisy Protected Attributescited by 17
  23. 2019Kernel Dependence Regularizers and Gaussian Processes with Applications to Algorithmic Fairnesscited by 16
  24. 2020Ensuring Fairness under Prior Probability Shiftscited by 16
  25. 2020Beyond H-Divergence: Domain Adaptation Theory With Jensen-Shannon Divergencecited by 15
  26. 2021Generalization Bounds for Noisy Iterative Algorithms Using Properties of Additive Noise Channelscited by 15
  27. 2020A Maximal Correlation Approach to Imposing Fairness in Machine Learningcited by 13
  28. 2022On the Generalization for Transfer Learning: An Information-Theoretic Analysiscited by 11
  29. 2022Attainability and Optimality: The Equalized Odds Fairness Revisitedcited by 10
  30. 2024Class-wise Generalization Error: an Information-Theoretic Analysiscited by 2
  31. 2024Fairness via Independence: A (Conditional) Distance Covariance Frameworkcited by 2

Built on by

Cited by year (2026 to 14 Oct)

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

3 papers.

  1. 2025FedPF: Accurate Target Privacy Preserving Federated Learning Balancing Fairness and Utilitycited by 0
  2. 2026PAC-Bayesian Generalization Guarantees for Fairness on Stochastic and Deterministic Classifierscited by 0
  3. 2026No-Free-Fairness: Fundamental Limits and Trade-offs in Learning Systemscited by 0