发表机构
School of Information and Communication Engineering, Chungbuk National University（忠北国立大学信息与通信工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
FedA2L通过基于模型分歧信号动态调整分层学习率，解决了去中心化联邦学习中统一学习率导致的收敛效率低问题，可提升收敛速度、减少通信开销，适配多种场景。
AI 中文摘要
具有异构设备和有限协调能力的去中心化智能系统越来越依赖去中心化联邦学习（Decentralized Federated Learning, DFL）。然而，由于使用统一学习率（Learning Rate, LR）忽略了各层的优化需求，DFL在数据异质性下存在收敛效率低的问题。基础层负责维持网络一致性，而专用层则适配本地数据特征，这会导致梯度冲突并在非独立同分布（non-IID）条件下降低性能。为解决这一根本矛盾，本研究提出FedA2L，一种基于模型分歧信号动态调整分层学习率的方法。通过利用局部更新强度和网络一致性约束，FedA2L可无缝集成到现有DFL协议中，无需额外通信或协调。在DFL算法、多种模型架构和数据集上的广泛评估表明，FedA2L的收敛速度比普通DFL快4.94倍，与基于调度器的基线相比通信轮次最多减少59%。此外，FedA2L对严重数据异质性、更大网络规模和稀疏拓扑结构具有鲁棒性，可降低通信开销，是边缘和物联网部署中资源受限或大规模分布式学习的通用优化工具。代码已发布在该httpsURL。
英文摘要
Decentralized intelligence systems with heterogeneous devices and limited coordination increasingly rely on decentralized federated learning (DFL). However, DFL suffers from convergence inefficiency under data heterogeneity due to the use of a uniform learning rate (LR) that ignores layer-specific optimization needs. Foundational layers are responsible for maintaining network consensus, while specialized layers adapt to local data characteristics, leading to conflicting gradients and degraded performance under non-IID conditions. To address this fundamental tension, this work introduces FedA2L, a method that dynamically adjusts layer-wise LRs based on model divergence signals. By leveraging local update intensity and network consensus constraints, FedA2L seamlessly integrates into existing DFL protocols without additional communication or coordination. Extensive evaluations across DFL algorithms, various model architectures, and datasets demonstrate that FedA2L achieves up to 4.94 times faster convergence than vanilla DFL and reduces communication rounds by up to 59% compared to scheduler-based baselines. Furthermore, FedA2L exhibits resilience to severe data heterogeneity, larger network sizes, and sparse topologies, reducing communication overhead and establishing it as a versatile optimization tool for resource-constrained or large-scale distributed learning in edge and IoT deployments. The code is released at https://github.com/nclabteam/FedA2L.
Journal refFuture Generation Computer Systems 186 (2026) 108743
DOI：10.1016/j.future.2026.108743
