提出累积效用公平性，解决间歇参与下客户端长期受益不均问题。
Cumulative Utility Parity for Fair Federated Learning under Intermittent Client Participation
- 基于参与机会评估长期收益公平性，而非每轮表现
- 在非独立同分布数据上显著提升长期代表性均衡度
- 适合资源不均、参与不稳定的现实联邦学习场景
真实联邦学习系统中，客户端参与具有间歇性、异构性，且常与数据特征或资源限制相关。现有公平性方法主要关注参与时的损失或准确率均等，隐含假设各客户端有相似贡献机会。但当参与本身不均时，此类目标可能导致间歇性客户端被系统性低估，即使每轮表现看似公平。本文提出累积效用公平性原则，衡量客户端在每次参与机会中获得的长期收益是否相当，而非每轮表现。为此引入可访问性归一化的累积效用，将不可避免的物理约束与可避免的算法偏见（调度与聚合）分离。在时间偏斜、非独立同分布的联邦基准测试中，该方法显著改善了长期代表性公平性，同时保持近乎完美的性能。
原文摘要 · Abstract (English)
In real-world federated learning (FL) systems, client participation is intermittent, heterogeneous, and often correlated with data characteristics or resource constraints. Existing fairness approaches in FL primarily focus on equalizing loss or accuracy conditional on participation, implicitly assuming that clients have comparable opportunities to contribute over time. However, when participation itself is uneven, these objectives can lead to systematic under-representation of intermittently available clients, even if per-round performance appears fair. We propose cumulative utility parity, a fairness principle that evaluates whether clients receive comparable long-term benefit per participation opportunity, rather than per training round. To operationalize this notion, we introduce availability-normalized cumulative utility, which disentangles unavoidable physical constraints from avoidable algorithmic bias arising from scheduling and aggregation. Experiments on temporally skewed, non-IID federated benchmarks demonstrate that our approach substantially improves long-term representation parity, while maintaining near-perfect performance.
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。