arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2302.08533cs.LGcs.DC
收藏

作为网络效应博弈的联邦学习

Federated Learning as a Network Effects Game

Carnegie Mellon University(卡内基梅隆大学);University of Minnesota(明尼苏达大学)

Shengyuan Hu, Dung Daniel Ngo, Shuran Zheng, Virginia Smith, Zhiwei Steven Wu

摘要:

联邦学习(FL)旨在促进客户群体之间的协作,以提高机器学习的准确性,而无需直接共享本地数据。虽然有丰富的文献设计联邦学习算法,大多数以前的作品隐含地假设,所有的客户端都愿意参与FL计划。在实践中,客户可能不会从加入FL中受益,特别是考虑到与隐私和计算等问题相关的潜在成本。在这项工作中,我们研究了客户在联邦学习中的动机,以帮助服务提供商设计更好的解决方案,并确保客户做出更好的决策。我们是第一个将FL中的客户行为建模为网络效应游戏的公司,每个客户的利益取决于也加入网络的其他客户。使用这种设置,我们分析客户的参与动态和特征的均衡,没有客户有动机改变他们的决定。具体来说,我们表明,在人口动态自然收敛到平衡,而不需要明确的干预。最后,我们提供了一个具有成本效益的支付方案,激励客户端达到理想的平衡时,初始网络是空的。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑