FairEnergy: Contribution-Based Fairness meets Energy Efficiency in Federated Learning

TL;DR

FairEnergy is a novel framework that balances energy efficiency and fairness in federated learning by considering client contributions, leading to improved accuracy and significant energy savings in wireless edge systems.

Contribution

It introduces a contribution-based fairness framework that jointly optimizes device selection, bandwidth, and compression, addressing energy and fairness in federated learning.

Findings

Achieves higher accuracy than baseline methods.

Reduces energy consumption by up to 79%.

Effectively handles non-IID data distributions.

Abstract

Federated learning (FL) enables collaborative model training across distributed devices while preserving data privacy. However, balancing energy efficiency and fair participation while ensuring high model accuracy remains challenging in wireless edge systems due to heterogeneous resources, unequal client contributions, and limited communication capacity. To address these challenges, we propose FairEnergy, a fairness-aware energy minimization framework that integrates a contribution score capturing both the magnitude of updates and their compression ratio into the joint optimization of device selection, bandwidth allocation, and compression level. The resulting mixed-integer non-convex problem is solved by relaxing binary selection variables and applying Lagrangian decomposition to handle global bandwidth coupling, followed by per-device subproblem optimization. Experiments on non-IID data show that FairEnergy achieves higher accuracy while reducing energy consumption by up to 79\% compared to baseline strategies.