Graph Federated Learning for CIoT Devices in Smart Home Applications

TL;DR

This paper introduces G-Fedfilt, a graph signal processing-inspired aggregation rule for federated learning in smart home CIoT devices, enhancing personalization, communication efficiency, and accuracy in heterogeneous environments.

Contribution

It proposes a novel graph filtering-based aggregation method, G-Fedfilt, that improves personalization and communication efficiency in federated learning for CIoT devices.

Findings

G-Fedfilt outperforms FedAvg by up to 3.99% in classification accuracy.

It enables intrinsic smooth clustering based on device connectivity.

The scheme reduces communication latency through adaptive time-scheduling.

Abstract

This paper deals with the problem of statistical and system heterogeneity in a cross-silo Federated Learning (FL) framework where there exist a limited number of Consumer Internet of Things (CIoT) devices in a smart building. We propose a novel Graph Signal Processing (GSP)-inspired aggregation rule based on graph filtering dubbed ``G-Fedfilt''. The proposed aggregator enables a structured flow of information based on the graph's topology. This behavior allows capturing the interconnection of CIoT devices and training domain-specific models. The embedded graph filter is equipped with a tunable parameter which enables a continuous trade-off between domain-agnostic and domain-specific FL. In the case of domain-agnostic, it forces G-Fedfilt to act similar to the conventional Federated Averaging (FedAvg) aggregation rule. The proposed G-Fedfilt also enables an intrinsic smooth clustering based on the graph connectivity without explicitly specified which further boosts the personalization of the models in the framework. In addition, the proposed scheme enjoys a communication-efficient time-scheduling to alleviate the system heterogeneity. This is accomplished by adaptively adjusting the amount of training data samples and sparsity of the models' gradients to reduce communication desynchronization and latency. Simulation results show that the proposed G-Fedfilt achieves up to $3.99\% $ better classification accuracy than the conventional FedAvg when concerning model personalization on the statistically heterogeneous local datasets, while it is capable of yielding up to $2.41\%$ higher accuracy than FedAvg in the case of testing the generalization of the models.