Over-the-Air Fair Federated Learning via Multi-Objective Optimization
Shayan Mohajer Hamidi, Ali Bereyhi, Saba Asaad, H. Vincent Poor
TL;DR
This work tackles fairness in federated learning under client data heterogeneity by formulating FL as a multi-objective optimization problem and solving it with a Modified Chebyshev scheme to adaptively weight client gradients. It extends the approach to a wireless setting via Over-the-Air Computation, deriving optimal transmit and de-noising scalars to enable unbiased, low-variance aggregation over a fading MAC. Key contributions include (i) the first fair FL algorithm compatible with OTA computation, (ii) a modified Chebyshev MoM for balancing fairness and average performance, (iii) closed-form solutions for transmit scalars and a PS de-noising scalar, and (iv) extensive experiments demonstrating improved fairness and robust performance compared to baselines. The proposed OTA-FFL framework offers a practical means to achieve fair, scalable model training in wireless federated learning environments, with clear implications for privacy-preserving and resource-constrained deployments.
Abstract
In federated learning (FL), heterogeneity among the local dataset distributions of clients can result in unsatisfactory performance for some, leading to an unfair model. To address this challenge, we propose an over-the-air fair federated learning algorithm (OTA-FFL), which leverages over-the-air computation to train fair FL models. By formulating FL as a multi-objective minimization problem, we introduce a modified Chebyshev approach to compute adaptive weighting coefficients for gradient aggregation in each communication round. To enable efficient aggregation over the multiple access channel, we derive analytical solutions for the optimal transmit scalars at the clients and the de-noising scalar at the parameter server. Extensive experiments demonstrate the superiority of OTA-FFL in achieving fairness and robust performance compared to existing methods.