Adaptive Federated Learning in Heterogeneous Wireless Networks with Independent Sampling
This work addresses efficiency challenges in federated learning for practical heterogeneous wireless networks, representing an incremental improvement over existing sampling methods.
The paper tackles the problem of minimizing wall-clock training time in federated learning over heterogeneous wireless networks by proposing an independent client sampling strategy that accounts for data and system heterogeneity. Experimental results show the proposed scheme substantially outperforms current best sampling schemes under various models and datasets.
Federated Learning (FL) algorithms commonly sample a random subset of clients to address the straggler issue and improve communication efficiency. While recent works have proposed various client sampling methods, they have limitations in joint system and data heterogeneity design, which may not align with practical heterogeneous wireless networks. In this work, we advocate a new independent client sampling strategy to minimize the wall-clock training time of FL, while considering data heterogeneity and system heterogeneity in both communication and computation. We first derive a new convergence bound for non-convex loss functions with independent client sampling and then propose an adaptive bandwidth allocation scheme. Furthermore, we propose an efficient independent client sampling algorithm based on the upper bounds on the convergence rounds and the expected per-round training time, to minimize the wall-clock time of FL, while considering both the data and system heterogeneity. Experimental results under practical wireless network settings with real-world prototype demonstrate that the proposed independent sampling scheme substantially outperforms the current best sampling schemes under various training models and datasets.