动态调整训练周期与批次，提升异构设备下联邦学习效率
Adaptive Deadline and Batch Layered Synchronized Federated Learning
- 根据设备能力动态设定每轮截止时间和批次大小
- 在限定时间内使模型收敛速度更快，最终准确率更高
- 适合资源差异大的边缘设备协同训练场景
联邦学习（FL）可在分布式边缘设备间协作训练模型并保护数据隐私，通常采用基于轮次的同步模式。然而，由于设备异构性，慢速设备（拖尾者）会导致同步延迟或降低全局更新质量。现有方法如固定截止时间、客户端选择和分层局部聚合虽能缓解此问题，但将轮次时长和本地计算量视为静态参数，在严格时间约束下效果受限。本文提出 ADEL-FL 框架，联合优化每轮截止时间和用户特定的分层聚合批次大小。通过构建约束优化问题，最小化总训练时间与全局轮次下期望的 L2 距离至全局最优。在指数计算模型下提供收敛性分析，证明 ADEL-FL 可产生无偏更新且方差有界。大量实验表明，在异构条件下，该方法在收敛速度与最终准确率上均优于其他方法。
原文摘要 · Abstract (English)
Federated learning (FL) enables collaborative model training across distributed edge devices while preserving data privacy, and typically operates in a round-based synchronous manner. However, synchronous FL suffers from latency bottlenecks due to device heterogeneity, where slower clients (stragglers) delay or degrade global updates. Prior solutions, such as fixed deadlines, client selection, and layer-wise partial aggregation, alleviate the effect of stragglers, but treat round timing and local workload as static parameters, limiting their effectiveness under strict time constraints. We propose ADEL-FL, a novel framework that jointly optimizes per-round deadlines and user-specific batch sizes for layer-wise aggregation. Our approach formulates a constrained optimization problem minimizing the expected L2 distance to the global optimum under total training time and global rounds. We provide a convergence analysis under exponential compute models and prove that ADEL-FL yields unbiased updates with bounded variance. Extensive experiments demonstrate that ADEL-FL outperforms alternative methods in both convergence rate and final accuracy under heterogeneous conditions.
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。