Yardstick-Stackelberg pricing-based incentive mechanism for Federated Learning in Edge Computing

IF 4.4 2区 计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE
Qianhui Yu , Hai Xue , Celimuge Wu , Ya Liu , Wunan Guo
{"title":"Yardstick-Stackelberg pricing-based incentive mechanism for Federated Learning in Edge Computing","authors":"Qianhui Yu ,&nbsp;Hai Xue ,&nbsp;Celimuge Wu ,&nbsp;Ya Liu ,&nbsp;Wunan Guo","doi":"10.1016/j.comnet.2025.111186","DOIUrl":null,"url":null,"abstract":"<div><div>Federated Learning (FL) enables collaborative model training across multiple participants without sharing original data, making it a valuable tool for preserving privacy in Mobile Edge Computing (MEC) environment. However, due to users’ varying levels of motivation and commitment, it is challenging to incentivize effective participation in FL. To address this, we propose a pricing-based incentive mechanism that enhances FL efficiency and energy sustainability in MEC. To be specific, we firstly develop the formula of incentive mechanism based on the yardstick pricing rule. Subsequently, we determine the optimal hyperparameters of the utility function aiming to maximize model accuracy. Additionally, we formulate a Stackelberg game to derive optimal reward strategies, balancing users’ transmission power allocation and the server’s reward distribution. Simulation results show that our proposed scheme outperforms other existing schemes with over 98.2% accuracy, 0.7% server utility enhancement, and 14.6% server loss decrease compared with static incentives. Moreover, our proposed scheme contributes to faster growth in both server and users utilities when compared with the advanced schemes by varying user numbers, which demonstrates its better scalability and adaptability.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":"262 ","pages":"Article 111186"},"PeriodicalIF":4.4000,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computer Networks","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1389128625001549","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE","Score":null,"Total":0}
引用次数: 0

Abstract

Federated Learning (FL) enables collaborative model training across multiple participants without sharing original data, making it a valuable tool for preserving privacy in Mobile Edge Computing (MEC) environment. However, due to users’ varying levels of motivation and commitment, it is challenging to incentivize effective participation in FL. To address this, we propose a pricing-based incentive mechanism that enhances FL efficiency and energy sustainability in MEC. To be specific, we firstly develop the formula of incentive mechanism based on the yardstick pricing rule. Subsequently, we determine the optimal hyperparameters of the utility function aiming to maximize model accuracy. Additionally, we formulate a Stackelberg game to derive optimal reward strategies, balancing users’ transmission power allocation and the server’s reward distribution. Simulation results show that our proposed scheme outperforms other existing schemes with over 98.2% accuracy, 0.7% server utility enhancement, and 14.6% server loss decrease compared with static incentives. Moreover, our proposed scheme contributes to faster growth in both server and users utilities when compared with the advanced schemes by varying user numbers, which demonstrates its better scalability and adaptability.
边缘计算中基于尺度- stackelberg定价的联邦学习激励机制
联邦学习(FL)可以在不共享原始数据的情况下跨多个参与者进行协作模型训练,使其成为在移动边缘计算(MEC)环境中保护隐私的宝贵工具。然而,由于用户的动机和承诺水平不同,激励有效参与FL是具有挑战性的。为了解决这个问题,我们提出了一种基于定价的激励机制,以提高MEC的FL效率和能源可持续性。具体而言,我们首先建立了基于标尺定价规则的激励机制公式。随后,我们确定了效用函数的最优超参数,以使模型精度最大化。此外,我们制定了一个Stackelberg博弈来获得最优的奖励策略,以平衡用户的传输功率分配和服务器的奖励分配。仿真结果表明,与静态激励方案相比,我们提出的方案的准确率超过98.2%,服务器利用率提高0.7%,服务器损耗降低14.6%。此外,通过改变用户数量,我们提出的方案与先进的方案相比,在服务器和用户效用方面都有更快的增长,这表明我们提出的方案具有更好的可扩展性和适应性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Computer Networks
Computer Networks 工程技术-电信学
CiteScore
10.80
自引率
3.60%
发文量
434
审稿时长
8.6 months
期刊介绍: Computer Networks is an international, archival journal providing a publication vehicle for complete coverage of all topics of interest to those involved in the computer communications networking area. The audience includes researchers, managers and operators of networks as well as designers and implementors. The Editorial Board will consider any material for publication that is of interest to those groups.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信