Towards Realistic Mechanisms That Incentivize Federated Participation and Contribution
This addresses the free-rider problem in federated learning for edge device settings, representing an incremental step towards more realistic implementations.
The paper tackles the problem of incentivizing edge device participation and data contribution in federated learning by proposing RealFM, a mechanism that realistically models device utility and provably removes the free-rider dilemma, improving device and server utility by over 3 and 4 magnitudes respectively on real-world data.
Edge device participation in federating learning (FL) is typically studied through the lens of device-server communication (e.g., device dropout) and assumes an undying desire from edge devices to participate in FL. As a result, current FL frameworks are flawed when implemented in realistic settings, with many encountering the free-rider dilemma. In a step to push FL towards realistic settings, we propose RealFM: the first federated mechanism that (1) realistically models device utility, (2) incentivizes data contribution and device participation, (3) provably removes the free-rider dilemma, and (4) relaxes assumptions on data homogeneity and data sharing. Compared to previous FL mechanisms, RealFM allows for a non-linear relationship between model accuracy and utility, which improves the utility gained by the server and participating devices. On real-world data, RealFM improves device and server utility, as well as data contribution, by over 3 and 4 magnitudes respectively compared to baselines.