Mitigating federated learning contribution allocation instability through randomized aggregation

Federated learning (FL) is a collaborative and privacy-preserving Machine Learning paradigm, allowing the development of robust models without the need to centralise sensitive data. A critical challenge in FL lies in fairly and accurately allocating contributions from diverse participants. Inaccurate allocation can undermine trust, lead to unfair compensation, and thus participants may lack the incentive to join or actively contribute to the federation. Various remuneration strategies have been proposed to date, including auction-based approaches and Shapley-value based methods, the latter offering a means to quantify the contribution of each participant. However, little to no work has studied the stability of these contribution evaluation methods. In this paper, we focus on calculating contributions using gradient-based model reconstruction techniques with Shapley values. We first show that baseline Shapley values do not accurately reflect clients' contributions, leading to unstable reward allocations amongst participants in a cross-silo federation. We then introduce \textsc{FedRandom}, a new method that mitigates these shortcomings with additional data samplings, and show its efficacy at increasing the stability of contribution evaluation in federated learning.