Personalized Over-the-Air Federated Learning with Personalized Reconfigurable Intelligent Surfaces
This addresses data heterogeneity in wireless federated learning for users with diverse datasets, though it appears incremental as it builds on existing personalized and over-the-air methods.
The paper tackles the problem of bandwidth-efficient federated learning with non-i.i.d. data by proposing a personalized over-the-air federated learning scheme using multi-task learning and personal reconfigurable intelligent surfaces, demonstrating improved performance on the Fashion-MNIST dataset.
Over-the-air federated learning (OTA-FL) provides bandwidth-efficient learning by leveraging the inherent superposition property of wireless channels. Personalized federated learning balances performance for users with diverse datasets, addressing real-life data heterogeneity. We propose the first personalized OTA-FL scheme through multi-task learning, assisted by personal reconfigurable intelligent surfaces (RIS) for each user. We take a cross-layer approach that optimizes communication and computation resources for global and personalized tasks in time-varying channels with imperfect channel state information, using multi-task learning for non-i.i.d data. Our PROAR-PFed algorithm adaptively designs power, local iterations, and RIS configurations. We present convergence analysis for non-convex objectives and demonstrate that PROAR-PFed outperforms state-of-the-art on the Fashion-MNIST dataset.