Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Personalized Over-the-Air Federated Learning with Personalized Reconfigurable Intelligent Surfaces

Jiayu Mao, Aylin Yener

TL;DR

This work introduces PROAR-PFed, the first personalized over-the-air federated learning framework that equips each client with a personal RIS to address non-i.i.d. data in time-varying wireless channels. It adopts a bi-level, multi-task objective combining local personalization and global learning, and solves a cross-layer optimization problem that jointly designs transmit power, local iterations, and RIS phase shifts under imperfect CSI. The authors provide convergence analysis for non-convex objectives and demonstrate, on Fashion-MNIST, that PROAR-PFed achieves superior personalized and global accuracy compared with state-of-the-art methods, particularly as data heterogeneity increases. The results highlight the practical value of per-user RIS and dynamic resource allocation for efficient and effective OTA-FL in next-generation wireless networks.

Abstract

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.

Personalized Over-the-Air Federated Learning with Personalized Reconfigurable Intelligent Surfaces

TL;DR

This work introduces PROAR-PFed, the first personalized over-the-air federated learning framework that equips each client with a personal RIS to address non-i.i.d. data in time-varying wireless channels. It adopts a bi-level, multi-task objective combining local personalization and global learning, and solves a cross-layer optimization problem that jointly designs transmit power, local iterations, and RIS phase shifts under imperfect CSI. The authors provide convergence analysis for non-convex objectives and demonstrate, on Fashion-MNIST, that PROAR-PFed achieves superior personalized and global accuracy compared with state-of-the-art methods, particularly as data heterogeneity increases. The results highlight the practical value of per-user RIS and dynamic resource allocation for efficient and effective OTA-FL in next-generation wireless networks.

Abstract

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.
Paper Structure (6 sections, 2 theorems, 15 equations, 2 figures, 1 table, 1 algorithm)

This paper contains 6 sections, 2 theorems, 15 equations, 2 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model
  3. Algorithm Design
  4. Convergence Analysis
  5. Numerical Results
  6. Conclusion

Key Result

theorem 1

With Assumptions a_smooth- a_bounded, a constant global learning rate $\eta_t = \eta \leq \frac{1}{L}$, a constant local learning rate $\eta_v \leq \frac{1}{\sqrt{2L^2 + 2 \lambda^2}}$ and $T\geq 4$, for each device $i\in [m]$, we have: where $\frac{1}{\bar{\beta}^2} = \frac{1}{T} \sum_{t=0}^{T-1} \frac{1}{\beta_t^2}$, $\tau_v = \tau_v^i$.

Figures (2)

  • Figure 1: The personal RIS-assisted communication system.
  • Figure 2: Average test accuracy for personalized tasks when $\gamma=0.5$.

Theorems & Definitions (3)

  • theorem 1
  • proof : Proof Highlights
  • corollary 1