Personalizing Federated Learning with Over-the-Air Computations
This work addresses efficiency and personalization challenges in federated learning for edge networks, offering an incremental improvement over existing methods.
The paper tackles communication bottlenecks and data heterogeneity in federated edge learning by introducing a distributed training paradigm using analog over-the-air computation and a bi-level optimization framework for personalization, resulting in enhanced generalization and robustness for client models with theoretical convergence analysis and experimental validation.
Federated edge learning is a promising technology to deploy intelligence at the edge of wireless networks in a privacy-preserving manner. Under such a setting, multiple clients collaboratively train a global generic model under the coordination of an edge server. But the training efficiency is often throttled by challenges arising from limited communication and data heterogeneity. This paper presents a distributed training paradigm that employs analog over-the-air computation to address the communication bottleneck. Additionally, we leverage a bi-level optimization framework to personalize the federated learning model so as to cope with the data heterogeneity issue. As a result, it enhances the generalization and robustness of each client's local model. We elaborate on the model training procedure and its advantages over conventional frameworks. We provide a convergence analysis that theoretically demonstrates the training efficiency. We also conduct extensive experiments to validate the efficacy of the proposed framework.