mlsanityData to 7 Oct 2026

The Impact of Differential Privacy on Group Disparity Mitigation

arXiv 2203.02745 ↗First posted 5 Mar 2022cs.CR cs.CL cs.LGcited by 11

Not on the map. Why?

Abstract (arXiv)

The performance cost of differential privacy has, for some applications, been shown to be higher for minority groups; fairness, conversely, has been shown to disproportionally compromise the privacy of members of such groups. Most work in this area has been restricted to computer vision and risk assessment. In this paper, we evaluate the impact of differential privacy on fairness across four tasks, focusing on how attempts to mitigate privacy violations and between-group performance differences interact: Does privacy inhibit attempts to ensure fairness? To this end, we train (varepsilon,δ)-differentially private models with empirical risk minimization and group distributionally robust training objectives. Consistent with previous findings, we find that differential privacy increases between-group performance differences in the baseline setting; but more interestingly, differential privacy reduces between-group performance differences in the robust setting. We explain this by reinterpreting differential privacy as regularization.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 31 of its 61 references that are in the map. First posted on arXiv 5 Mar 2022.

Builds on

31 of 61 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2018GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understandingcited by 4,854
  4. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,243
  5. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  6. 2016Deep Learning with Differential Privacycited by 2,682
  7. 2019Invariant Risk Minimizationcited by 1,832
  8. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,366
  9. 2017Renyi Differential Privacycited by 643
  10. 2018A Reductions Approach to Fair Classificationcited by 585
  11. 2019Differential Privacy Has Disparate Impact on Model Accuracycited by 316
  12. 2020An Investigation of Why Overparameterization Exacerbates Spurious Correlationscited by 310
19 more
  1. 2020Environment Inference for Invariant Learningcited by 306
  2. 2020Gradient Starvation: A Learning Proclivity in Neural Networkscited by 262
  3. 2018Empirical Risk Minimization under Fairness Constraintscited by 226
  4. 2015Cost Sensitive Learning of Deep Feature Representations from Imbalanced Datacited by 167
  5. 2019Distributionally Robust Language Modelingcited by 142
  6. 2015Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networkscited by 115
  7. 2019Fairness risk measurescited by 85
  8. 2018A General Approach to Adding Differential Privacy to Iterative Training Procedurescited by 84
  9. 2020On the Privacy Risks of Algorithmic Fairnesscited by 82
  10. 2018Differentially Private Fair Learningcited by 76
  11. 2020Neither Private Nor Fair: Impact of Data Imbalance on Utility and Fairness in Differential Privacycited by 61
  12. 2020Differentially Private and Fair Deep Learning: A Lagrangian Dual Approachcited by 53
  13. 2019Leveraging Hierarchical Representations for Preserving Privacy and Utility in Textcited by 49
  14. 2020Differentially Private Language Models Benefit from Public Pre-trainingcited by 46
  15. 2019Evaluating Lottery Tickets Under Distributional Shiftscited by 23
  16. 2021Investigating Trade-offs in Utility, Fairness and Differential Privacy in Neural Networkscited by 17
  17. 2020How to Democratise and Protect AI: Fair and Differentially Private Decentralised Deep Learningcited by 12
  18. 2015Cost-aware Pre-training for Multiclass Cost-sensitive Deep Learningcited by 9
  19. 2021Assessing Fairness in Classification Parity of Machine Learning Models in Healthcarecited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 3; 2024: 1; 2025: 2; 2026: 4.

11 papers.

  1. 2022Individual Privacy Accounting for Differentially Private Stochastic Gradient Descentcited by 20
  2. 2023DP-NMT: Scalable Differentially-Private Machine Translationcited by 7
  3. 2023Trade-Offs Between Fairness and Privacy in Language Modelingcited by 4
  4. 2025LLM Bias Detection and Mitigation through the Lens of Desired Distributionscited by 4
  5. 2023Differential Privacy, Linguistic Fairness, and Training Data Influence: Impossibility and Possibility Theorems for Multilingual Language Modelscited by 1
  6. 2025Private and Fair Machine Learning: Revisiting the Disparate Impact of Differentially Private SGDcited by 1
  7. 2024Characterizing Stereotypical Bias from Privacy-preserving Pre-Trainingcited by 0
  8. 2026Private and interpretable clinical prediction with quantum-inspired tensor train modelscited by 0
  9. 2026Balancing Fairness, Privacy, and Accuracy: A Multitask Adversarial Framework for Centralized Data-Driven Systemscited by 0
  10. 2026Auditing Fairness-Privacy Trade-offs: Subpopulation-Level Effects of Fairness-Enhancing Algorithmscited by 0
  11. 2026The Privacy-Hallucination Tradeoff in Differentially Private Language Modelscited by 0