FedCure: Mitigating Participation Bias in Semi-Asynchronous Federated Learning with Non-IID Data
Yue Chen, Jianfeng Lu, Shuqing Cao, Wei Wang, Gang Li, Guanghui Wen
TL;DR
FedCure tackles participation bias in SAFL under non-IID data within a cloud-edge-client hierarchy by introducing a three-rule framework: a preference rule that forms stable, data-distribution-aware coalitions to minimize $\mathcal{\overline{JS}}$, a scheduling rule that couples Bayesian coalition dynamics with a virtual-queue Lyapunov approach to balance participation and efficiency, and a resource-allocation rule that optimizes CPU frequencies to meet delay targets. The authors prove the coalition game is an exact potential game and that the scheduling queues achieve mean-rate stability, deriving bounds on the efficiency-balance trade-off. Empirically, FedCure yields up to 5.1x accuracy gains and the lowest per-round latency variation across four real-world datasets, while maintaining long-term balance under diverse settings. The work provides a principled, scalable framework for mitigating participation bias in hierarchical SAFL with non-IID data and offers practical guidance for implementing coalition-based scheduling and resource control.
Abstract
While semi-asynchronous federated learning (SAFL) combines the efficiency of synchronous training with the flexibility of asynchronous updates, it inherently suffers from participation bias, which is further exacerbated by non-IID data distributions. More importantly, hierarchical architecture shifts participation from individual clients to client groups, thereby further intensifying this issue. Despite notable advancements in SAFL research, most existing works still focus on conventional cloud-end architectures while largely overlooking the critical impact of non-IID data on scheduling across the cloud-edge-client hierarchy. To tackle these challenges, we propose FedCure, an innovative semi-asynchronous Federated learning framework that leverages coalition construction and participation-aware scheduling to mitigate participation bias with non-IID data. Specifically, FedCure operates through three key rules: (1) a preference rule that optimizes coalition formation by maximizing collective benefits and establishing theoretically stable partitions to reduce non-IID-induced performance degradation; (2) a scheduling rule that integrates the virtual queue technique with Bayesian-estimated coalition dynamics, mitigating efficiency loss while ensuring mean rate stability; and (3) a resource allocation rule that enhances computational efficiency by optimizing client CPU frequencies based on estimated coalition dynamics while satisfying delay requirements. Comprehensive experiments on four real-world datasets demonstrate that FedCure improves accuracy by up to 5.1x compared with four state-of-the-art baselines, while significantly enhancing efficiency with the lowest coefficient of variation 0.0223 for per-round latency and maintaining long-term balance across diverse scenarios.