《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (11): 3551-3558.DOI: 10.11772/j.issn.1001-9081.2022111727

• 网络与通信 • 上一篇    下一篇

基于多领导者Stackelberg博弈的分层联邦学习激励机制设计

耿方兴1,2, 李卓1,2(), 陈昕2   

  1. 1.网络文化与数字传播北京市重点实验室(北京信息科技大学),北京 100101
    2.北京信息科技大学 计算机学院,北京 100101
  • 收稿日期:2022-11-21 修回日期:2023-04-03 接受日期:2023-04-04 发布日期:2023-05-08 出版日期:2023-11-10
  • 通讯作者: 李卓
  • 作者简介:耿方兴(1999—),男,河南驻马店人,硕士研究生,主要研究方向:边缘计算
    李卓(1983—),男,河南南阳人,副教授,博士,CCF会员,主要研究方向:移动无线网络、分布式计算 lizhuo@bistu.edu.cn
    陈昕(1965—),男,江西南昌人,教授,博士,CCF会员,主要研究方向:网络性能评价、网络安全。
  • 基金资助:
    北京市自然科学基金资助项目(4232024);国家重点研发计划项目(2022YFF0604502);国家自然科学基金资助项目(61872044);北京市青年拔尖人才项目

Incentive mechanism design for hierarchical federated learning based on multi-leader Stackelberg game

Fangxing GENG1,2, Zhuo LI1,2(), Xin CHEN2   

  1. 1.Beijing Key Laboratory of Internet Culture and Digital Dissemination Research (Beijing Information Science and Technology University),Beijing 100101,China
    2.Computer School,Beijing Information Science and Technology University,Beijing 100101,China
  • Received:2022-11-21 Revised:2023-04-03 Accepted:2023-04-04 Online:2023-05-08 Published:2023-11-10
  • Contact: Zhuo LI
  • About author:GENG Fangxing, born in 1999, M. S. candidate. His research interests include edge computing.
    LI Zhuo, born in 1983, Ph. D., associate professor. His research interests include mobile wireless network, distributed computing.
    CHEN Xin, born in 1965, Ph. D., professor. His research interests include network performance evaluation, network security.
  • Supported by:
    Beijing Natural Science Foundation(4232024);National Key Research and Development Program of China(2022YFF0604502);National Natural Science Foundation of China(61872044);Beijing Municipal Program for Young Talents

摘要:

分层联邦学习中隐私安全与资源消耗等问题的存在降低了参与者的积极性。为鼓励足够多的参与者积极参与学习任务,并针对多移动设备与多边缘服务器之间的决策问题,提出基于多领导者Stackelberg博弈的激励机制。首先,通过量化移动设备的成本效用与边缘服务器的支付报酬,构建效用函数并定义最优化问题;其次,将移动设备之间的交互建模为演化博弈,将边缘服务器之间的交互建模为非合作博弈。为求解最优边缘服务器选择和定价策略,提出多轮迭代边缘服务器选择算法(MIES)和梯度迭代定价算法(GIPA),前者用于求解移动设备之间的演化博弈均衡解,后者用于求解边缘服务器之间的定价竞争问题。实验结果表明,所提算法GIPA与最优定价预测策略(OPPS)、历史最优定价策略(HOPS)和随机定价策略(RPS)相比,可使边缘服务器的平均效用分别提高4.06%、10.08%和31.39%。

关键词: 分层联邦学习, 激励机制, 定价策略, 多领导者Stackelberg博弈, 演化博弈

Abstract:

The existence of privacy security and resource consumption issues in hierarchical federated learning reduces the enthusiasm of participants. To encourage a sufficient number of participants to actively participate in learning tasks and address the decision-making problem between multiple mobile devices and multiple edge servers, an incentive mechanism based on multi-leader Stackelberg game was proposed. Firstly, by quantifying the cost-utility of mobile devices and the payment of edge servers, a utility function was constructed, and an optimization problem was defined. Then, the interaction among mobile devices was modeled as an evolutionary game, and the interaction among edge servers was modeled as a non-cooperative game. To solve the optimal edge server selection and pricing strategy, a Multi-round Iterative Edge Server selection algorithm (MIES) and a Gradient Iterative Pricing Algorithm (GIPA) were proposed. The former was used to solve the evolutionary game equilibrium solution among mobile devices, and the latter was used to solve the pricing competition problem among edge servers. Experimental results show that compared with Optimal Pricing Prediction Strategy (OPPS), Historical Optimal Pricing Strategy (HOPS) and Random Pricing Strategy (RPS), GIPA can increase the average utility of edge servers by 4.06%, 10.08%, and 31.39% respectively.

Key words: hierarchical federated learning, incentive mechanism, pricing strategy, multi-leader Stackelberg game, evolutionary game

中图分类号: