Hybrid IoT Device Selection With Knowledge Transfer for Federated Learning

Federated learning (FL) enables collaborative model training across massively distributed edge devices, such as Internet of Things (IoT) nodes. However, resource constraints impose a major challenge, as there exists a tradeoff between maximizing learning accuracy and minimizing communication overhead between the resource-limited devices. In this article, we present a device selection approach for heterogeneous FL systems based on multiobjective optimization (MOO) and knowledge transfer (KT). We formulate the resource constraint in federated optimization as a multiobjective problem, and obtain Pareto-optimal solutions balancing resource efficiency and test accuracy. Additionally, we introduce an innovative KT mechanism that propagates the globally optimal models obtained during MOO to subsequent FL tasks, further expediting convergence. The multiobjective formulation and KT provide new insights into efficient and robust FL for resource-constrained IoT applications. We conduct extensive experiments on real-world data sets. Results demonstrate that our method achieves up to 11% higher accuracy than state-of-the-art methods, while effectively mitigating resource constraints. Impact Statement—FL is an efficient algorithm that enables everything to be interconnected without sharing data. However, resource constraint is the main challenge for federated optimization problems. Although many works have proposed various solutions from different perspectives, these methods cannot simultaneously minimize the communication resource cost while ensuring algorithm performance. We propose an automatic device selection algorithm for federated systems based on MOO and KT. This work not only reduces the global resource usage rate of FL but also enables it to converge quickly.