Federated learning (FL) facilitates new applications at the edge, especially for wearable and Internet-of-Thing devices. Such devices capture a large and diverse amount of data, but they have memory, compute, power, and connectivity constraints which hinder their participation in FL. We propose Centaur, a multitier FL framework, enabling ultra-constrained devices to efficiently participate in FL on large neural nets. Centaur combines two major ideas: (i) a data selection scheme to choose a portion of samples that accelerates the learning, and (ii) a partition-based training algorithm that integrates both constrained and powerful devices owned by the same user. Evaluations, on four benchmark neural nets and three datasets, show that Centaur gains ~10\% higher accuracy than local training on constrained devices with ~58\% energy saving on average. Our experimental results also demonstrate the superior efficiency of Centaur when dealing with imbalanced data, client participation heterogeneity, and various network connection probabilities.

翻译：摘要：联邦学习（FL）推动了边缘设备上新型应用的发展，尤其是可穿戴设备和物联网设备。这些设备捕获海量且多样化的数据，但其内存、计算、功耗和连接性限制阻碍了它们参与FL。我们提出了一种多层FL框架Centaur，使超受限设备能够高效参与大型神经网络的联邦学习。Centaur融合了两项核心思想：（i）一种数据选择方案，用于选取部分样本以加速学习进程；（ii）一种基于分区的训练算法，整合同一用户拥有的受限设备与高性能设备。在四个基准神经网络和三个数据集上的评估表明，与在受限设备上进行本地训练相比，Centaur的准确率提升约10%，平均节能约58%。我们的实验结果还证明了Centaur在处理不平衡数据、客户端参与异质性以及不同网络连接概率时的卓越效率。