SlimFL: Federated Learning with Superposition Coding over Slimmable Neural Networks

TL;DR

SlimFL introduces a federated learning framework that combines superposition coding with slimmable neural networks to enhance communication efficiency and robustness against data heterogeneity and channel variability.

Contribution

It proposes a novel FL framework integrating superposition coding and slimmable neural networks, addressing communication, energy efficiency, and data heterogeneity challenges.

Findings

SlimFL achieves improved communication efficiency.

It effectively handles non-IID data distributions.

The framework demonstrates robustness under poor channel conditions.

Abstract

Federated learning (FL) is a key enabler for efficient communication and computing, leveraging devices' distributed computing capabilities. However, applying FL in practice is challenging due to the local devices' heterogeneous energy, wireless channel conditions, and non-independently and identically distributed (non-IID) data distributions. To cope with these issues, this paper proposes a novel learning framework by integrating FL and width-adjustable slimmable neural networks (SNN). Integrating FL with SNNs is challenging due to time-varying channel conditions and data distributions. In addition, existing multi-width SNN training algorithms are sensitive to the data distributions across devices, which makes SNN ill-suited for FL. Motivated by this, we propose a communication and energy-efficient SNN-based FL (named SlimFL) that jointly utilizes superposition coding (SC) for global model aggregation and superposition training (ST) for updating local models. By applying SC, SlimFL exchanges the superposition of multiple-width configurations decoded as many times as possible for a given communication throughput. Leveraging ST, SlimFL aligns the forward propagation of different width configurations while avoiding inter-width interference during backpropagation. We formally prove the convergence of SlimFL. The result reveals that SlimFL is not only communication-efficient but also deals with non-IID data distributions and poor channel conditions, which is also corroborated by data-intensive simulations.