Federated Learning (FL) aims to foster collaboration among a population of clients to improve the accuracy of machine learning without directly sharing local data. Although there has been rich literature on designing federated learning algorithms, most prior works implicitly assume that all clients are willing to participate in a FL scheme. In practice, clients may not benefit from joining in FL, especially in light of potential costs related to issues such as privacy and computation. In this work, we study the clients' incentives in federated learning to help the service provider design better solutions and ensure clients make better decisions. We are the first to model clients' behaviors in FL as a network effects game, where each client's benefit depends on other clients who also join the network. Using this setup we analyze the dynamics of clients' participation and characterize the equilibrium, where no client has incentives to alter their decision. Specifically, we show that dynamics in the population naturally converge to equilibrium without needing explicit interventions. Finally, we provide a cost-efficient payment scheme that incentivizes clients to reach a desired equilibrium when the initial network is empty.
[1]
Yae Jee Cho,et al.
To Federate or Not To Federate: Incentivizing Client Participation in Federated Learning
,
2022,
ArXiv.
[2]
Jon Kleinberg,et al.
Models of fairness in federated learning
,
2021,
ArXiv.
[3]
Jon Kleinberg,et al.
Optimality and Stability in Federated Learning: A Game-theoretic Approach
,
2021,
NeurIPS.
[4]
Avrim Blum,et al.
One for One, or All for All: Equilibria and Optimality of Collaboration in Federated Learning
,
2021,
ICML.
[5]
Virginia Smith,et al.
Ditto: Fair and Robust Federated Learning Through Personalization
,
2020,
ICML.
[6]
Jon Kleinberg,et al.
Model-sharing Games: Analyzing Federated Learning Under Voluntary Participation
,
2020,
AAAI.
[7]
Ariel D. Procaccia,et al.
Collaborative PAC Learning
,
2017,
NIPS.
[8]
Chris Arney,et al.
Networks, Crowds, and Markets: Reasoning about a Highly Connected World (Easley, D. and Kleinberg, J.; 2010) [Book Review]
,
2013,
IEEE Technology and Society Magazine.
[9]
C. Shapiro,et al.
Network Externalities, Competition, and Compatibility
,
1985
.