Designing Incentive Mechanisms for Fair Participation in Federated Learning.

Federated learning offers a collaborative method for training machine learning models using distributed data sources. Given its reliance on diverse contributions from multiple participants, equitable representation and rewards are paramount to its success. Upholding fairness is crucial to encourage active participation in this collaborative process. This paper presents a comprehensive analysis of existing mechanisms designed to incentivise fair engagement in federated learning, structured around a taxonomy aligned with the evolving dynamics of the federated learning sources. The study offers insights into cultivating environments that prioritise fairness and broad participation while suggesting avenues for future research.