提出相对公平的联邦学习框架，减少不同客户端间性能差异。
Federated Learning with Relative Fairness
- 用极小极大优化方法最小化客户端间的相对不公平性
- 引入基于损失比的公平指数，有效评估性能差距
- 算法兼顾通信与计算效率，适合实际部署
本文提出一种实现客户端相对公平性的联邦学习框架。传统方法通过保证所有客户端子组的最低性能来实现绝对公平，但忽略了子组间模型性能的差异。新框架采用极小极大问题方法，最小化相对不公平性，扩展了分布鲁棒优化（DRO）方法。提出一种新的公平指数，基于客户端间大损失与小损失的比率，用于评估和提升模型的相对公平性。理论证明该框架能持续降低不公平性。我们还设计了名为 extsc{Scaff-PD-IA} 的算法，在保持极小极大最优收敛率的同时平衡通信与计算效率。在真实数据集上的实证评估表明，该框架在维持模型性能的同时有效减少了子组间的性能差异。
原文摘要 · Abstract (English)
This paper proposes a federated learning framework designed to achieve \textit{relative fairness} for clients. Traditional federated learning frameworks typically ensure absolute fairness by guaranteeing minimum performance across all client subgroups. However, this approach overlooks disparities in model performance between subgroups. The proposed framework uses a minimax problem approach to minimize relative unfairness, extending previous methods in distributionally robust optimization (DRO). A novel fairness index, based on the ratio between large and small losses among clients, is introduced, allowing the framework to assess and improve the relative fairness of trained models. Theoretical guarantees demonstrate that the framework consistently reduces unfairness. We also develop an algorithm, named \textsc{Scaff-PD-IA}, which balances communication and computational efficiency while maintaining minimax-optimal convergence rates. Empirical evaluations on real-world datasets confirm its effectiveness in maintaining model performance while reducing disparity.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。