面向微控制器的去中心化联邦学习框架，支持低功耗设备高效协同训练。
Decentralised Resource Sharing in TinyML: Wireless Bilayer Gossip Parallel SGD for Collaborative Learning
- 分层通信：基于聚类的组内组间双层消息传播机制
- 在非独立同分布数据下仅损失8%准确率，收敛速度接近中心化方案
- 适用于资源受限、网络不稳定的边缘设备协同学习场景
随着微控制器（MCU）计算能力的提升，边缘设备已可支持机器学习模型。然而，在此类设备上部署去中心化联邦学习（DFL）面临连接不稳定、通信范围有限和网络拓扑动态变化等挑战。本文提出一种新框架——双层随机梯度下降（GD PSGD），通过分布式K均值（DKmeans）聚类实现地理分组，并结合八卦协议在组内与组间进行高效模型聚合。我们在CIFAR-10数据集上以MCUNet模型对比了中心化联邦学习（CFL）基准。结果表明，在独立同分布（IID）条件下，该方法精度与CFL相当，仅需额外1.8轮即可收敛；在非独立同分布（Non-IID）条件下，准确率损失低于8%，且在中等数据不平衡情况下表现稳定。该框架为资源受限环境下的可扩展、隐私保护学习提供了可行方案。
原文摘要 · Abstract (English)
With the growing computational capabilities of microcontroller units (MCUs), edge devices can now support machine learning models. However, deploying decentralised federated learning (DFL) on such devices presents key challenges, including intermittent connectivity, limited communication range, and dynamic network topologies. This paper proposes a novel framework, bilayer Gossip Decentralised Parallel Stochastic Gradient Descent (GD PSGD), designed to address these issues in resource-constrained environments. The framework incorporates a hierarchical communication structure using Distributed Kmeans (DKmeans) clustering for geographic grouping and a gossip protocol for efficient model aggregation across two layers: intra-cluster and inter-cluster. We evaluate the framework's performance against the Centralised Federated Learning (CFL) baseline using the MCUNet model on the CIFAR-10 dataset under IID and Non-IID conditions. Results demonstrate that the proposed method achieves comparable accuracy to CFL on IID datasets, requiring only 1.8 additional rounds for convergence. On Non-IID datasets, the accuracy loss remains under 8\% for moderate data imbalance. These findings highlight the framework's potential to support scalable and privacy-preserving learning on edge devices with minimal performance trade-offs.
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。