Benefits of the federation? Analyzing the impact of fair federated learning at the client level

Abstract

Federated Learning (FL) enables collaborative model training while preserving participating clients’ local data privacy. However, the diverse data distributions across different clients can exacerbate fairness issues, as biases inherent in client data may propagate across the federation. Although various approaches have been proposed to enhance fairness in FL, they typically focus on mitigating the bias of a single binary-sensitive attribute. This narrow focus often overlooks the complexity introduced by clients with conflicting or diverse fairness objectives. Such clients may contribute to the federation without experiencing any improvement in their own model’s performance or fairness regarding their specific sensitive attributes. In this paper, we compare three approaches to mitigate model unfairness in scenarios where clients have differing and potentially conflicting fairness requirements. By analysing disparities across sensitive attributes and model performance, we investigate the conditions under which clients benefit from federation participation. Our findings emphasise the importance of aligning federation objectives with diverse client needs to enhance participation and equitable outcomes in FL settings.