Fairness in Federated Learning: Fairness for Whom?
Not on the map. Why?
Abstract (arXiv)
Fairness in federated learning has emerged as a rapidly growing area of research, with numerous works proposing formal definitions and algorithmic interventions. Yet, despite this technical progress, fairness in FL is often defined and evaluated in ways that abstract away from the sociotechnical contexts in which these systems are deployed. In this paper, we argue that existing approaches tend to optimize narrow system level metrics, such as performance parity or contribution-based rewards, while overlooking how harms arise throughout the FL lifecycle and how they impact diverse stakeholders. We support this claim through a critical analysis of the literature, based on a systematic annotation of papers for their fairness definitions, design decisions, evaluation practices, and motivating use cases. Our analysis reveals five recurring pitfalls: 1) fairness framed solely through the lens of server client architecture, 2) a mismatch between simulations and motivating use-cases and contexts, 3) definitions that conflate protecting the system with protecting its users, 4) interventions that target isolated stages of the lifecycle while neglecting upstream and downstream effects, 5) and a lack of multi-stakeholder alignment where multiple fairness definitions can be relevant at once. Building on these insights, we propose a harm centered framework that links fairness definitions to concrete risks and stakeholder vulnerabilities. We conclude with recommendations for more holistic, context-aware, and accountable fairness research in FL.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 50 of its 94 references that are in the map. First posted on arXiv 27 May 2025.
Builds on
50 of 94 references.
References · 50
- 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
- 2014Deep Learning Face Attributes in the Wildcited by 3,693
- 2019Advances and Open Problems in Federated Learningcited by 3,046
- 2016Federated Learning: Strategies for Improving Communication Efficiencycited by 1,475
- 2019Towards Federated Learning at Scale: System Designcited by 832
- 2023Bias and Fairness in Large Language Models: A Surveycited by 639
- 2018Federated Learning for Mobile Keyboard Predictioncited by 572
- 2019Clustered Federated Learning: Model-Agnostic Distributed Multi-Task Optimization under Privacy Constraintscited by 503
- 2019Agnostic Federated Learningcited by 456
- 2021Retiring Adult: New Datasets for Fair Machine Learningcited by 449
- 2019Fair Resource Allocation in Federated Learningcited by 394
- 2018Client Selection for Federated Learning with Heterogeneous Resources in Mobile Edgecited by 352
38 more
- 2019Differential Privacy Has Disparate Impact on Model Accuracycited by 316
- 2019A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cyclecited by 193
- 2019Towards Fairer Datasets: Filtering and Balancing the Distribution of the People Subtree in the ImageNet Hierarchycited by 156
- 2023Debiasing Vision-Language Models via Biased Promptscited by 149
- 2020Algorithmic Fairnesscited by 143
- 2022Power to the People? Opportunities and Challenges for Participatory AIcited by 138
- 2022Sociotechnical Harms of Algorithmic Systems: Scoping a Taxonomy for Harm Reductioncited by 138
- 2021FedScale: Benchmarking Model and System Performance of Federated Learning at Scalecited by 136
- 2022FLamby: Datasets and Benchmarks for Cross-Silo Federated Learning in Realistic Healthcare Settingscited by 125
- 2021FairFed: Enabling Group Fairness in Federated Learningcited by 118
- 2020Collaborative Fairness in Federated Learningcited by 95
- 2019Problem Formulation and Fairnesscited by 85
- 2019Active Federated Learningcited by 63
- 2020An Efficiency-boosting Client Selection Scheme for Federated Learning with Fairness Guaranteecited by 62
- 2020Mitigating Bias in Federated Learningcited by 56
- 2018How Do Fairness Definitions Fare? Examining Public Attitudes Towards Algorithmic Definitions of Fairnesscited by 48
- 2019Energy Demand Prediction with Federated Learning for Electric Vehicle Networkscited by 46
- 2018Enhancing Person-Job Fit for Talent Recruitment: An Ability-aware Neural Network Approachcited by 39
- 2021Enforcing fairness in private federated learning via the modified method of differential multiplierscited by 38
- 2020Understanding Unintended Memorization in Federated Learningcited by 36
- 2020Federated Edge Learning : Design Issues and Challengescited by 26
- 2023Privacy and Fairness in Federated Learning: on the Perspective of Trade-offcited by 26
- 2022Minimax Demographic Group Fairness in Federated Learningcited by 24
- 2024A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harmscited by 19
- 2022PrivFairFL: Privacy-Preserving Group Fairness in Federated Learningcited by 13
- 2021The Contestation of Tech Ethics: A Sociotechnical Approach to Technology Ethics in Practicecited by 12
- 2023Fairness and Privacy-Preserving in Federated Learning: A Surveycited by 12
- 2024Lazy Data Practices Harm Fairness Researchcited by 10
- 2022FOCUS: Fairness via Agent-Awareness for Federated Learning on Heterogeneous Datacited by 6
- 2022Multi-Layer Personalized Federated Learning for Mitigating Biases in Student Predictive Analyticscited by 5
- 2023Mitigating Group Bias in Federated Learning for Heterogeneous Devicescited by 5
- 2024The Case for Globalizing Fairness: A Mixed Methods Study on Colonialism, AI, and Health in Africacited by 5
- 2024Achieving Fairness Across Local and Global Models in Federated Learningcited by 5
- 2023Fairness and Privacy in Federated Learning and Their Implications in Healthcarecited by 3
- 2025Enhancing Privacy in the Early Detection of Sexual Predators Through Federated Learning and Differential Privacycited by 2
- 2024Mitigating System Bias in Resource Constrained Asynchronous Federated Learning Systemscited by 1
- 2024Boosting Fairness and Robustness in Over-the-Air Federated Learningcited by 1
- 2024Privacy-preserving gradient-based fair federated learningcited by 1
Built on by
Citing papers per year: 2025: 1.
1 paper.