计算机科学 ›› 2024, Vol. 51 ›› Issue (3): 20-29.doi: 10.11896/jsjkx.230700194
王鑫1,2, 黄伟口1, 孙凌云2
WANG Xin1,2, HUANG Weikou1, SUN Lingyun2
摘要: 联邦学习作为一种分布式机器学习,有效地解决了大数据时代的数据共享难题。其中,跨机构联邦学习是机构之间互相合作的一种联邦学习类型。如何在跨机构合作的过程中设计合理的激励机制十分重要。文中从跨机构合作的角度,对现有的跨机构联邦学习的激励机制研究进行了综述。首先介绍跨机构合作过程中的3个基本问题,即高隐私性、数据异质性、公平性,然后分析了以全局模型为中心和以参与者为中心这两种不同的跨机构合作模式下的激励机制设计方法,最后总结了影响跨机构合作稳定发展的几个影响因素,即参与者的数据演变、参与者合作关系变动和参与者的负面行为,并展望了跨机构联邦合作的未来方向。
中图分类号:
[1]MCMAHAN B,MOORE E,RAMAGE D,et al.Communica-tion-efficient learning of deep networks from decentralized data [C]//Artificial Intelligence and Statistics.PMLR,2017:1273-1282. [2]NIE W,XIN L,LI Z.Study on the Application of FederatedLearning in Medical Informatization[J]. Journal of Medical Informatics,2022,43(10):12-13. [3]ZHAN Y,ZHANG J,HONG Z,et al.A survey of incentivemechanism design for federated learning[J].IEEE Transactions on Emerging Topics in Computing,2021,10(2):1035-1044. [4]ZENG R,ZENG C,WANG X,et al.Incentive Mechanisms in Federated Learning and A Game-Theoretical Approach [J].IEEE Network,2022,36(6):229-235. [5]LIANG W,LIU B,LIN W,et al.Survey of Incentive Mechanism for Federated Learning [J].Computer Science,2022,49(12):46-52. [6]TU X,ZHU K,LUONG N,et al.Incentive mechanisms for fe-derated learning:From economic and game theoretic perspective [J].IEEE Transactions on Cognitive Communications and Networking,2022,8(3):1566-1593. [7]WITT L,HEYER M,TOYODA K,et al.Decentral and In-centivized Federated Learning Frameworks:A Systematic Literature Review [J].IEEE Internet of Things Journal,2023,10(4):3642-3663. [8]LIU Y,KANG Y,ZOU T,et al.Vertical federated learning [J].arXiv:2211.12814,2022. [9]YAN Y,LIGETI P.A Survey of Personalized and IncentiveMechanisms for Federated Learning [C]//2022 IEEE 2nd Conference on Information Technology and Data Science.IEEE,2022:324-329. [10]HUANG C,HUANG J,LIU X.Cross-silo federated learning:Challenges and opportunities [J].arXiv:2206.12949,2022. [11]DWORK C.Differential privacy [C]//International Colloquium on Automata,Languages,and Programming.Berlin,Heidelberg:Springer,2006:1-12. [12]WANG C,WU X,LIU G,et al.Safeguarding cross-silo federated learning with local differential privacy [J].Digital Communications and Networks,2022,8(4):446-454. [13]REN K.Differentially Private Auction for Federated Learning with Non-IID Data [C]//2022 International Conference on Service Science.IEEE,2022:305-312. [14]ZHANG L,ZHU T,XIONG P,et al.A Game-theoretic Federated Learning Framework for Data Quality Improvement [J].IEEE Transactions on Knowledge and Data Engineering,2023,35(11):10952-10966. [15]WANG D,REN J,WANG Z,et al.PrivAim:A Dual-PrivacyPreserving and Quality-Aware Incentive Mechanism for Federated Learning [J].IEEE Transactions on Computers,2023,72(7):1913-1927. [16]WEI K,LI J,DING M,et al.Federated learning with differential privacy:Algorithms and performance analysis[J].IEEE Trans-actions on Information Forensics and Security,2020(15):3454-3469. [17]LIU K,HU S,WU S Z,et al.On privacy and personalization in cross-silo federated learning [J].Advances in Neural Information Processing Systems,2022,35:5925-5940. [18]ZHENG S,CAO Y,YOSHIKAWA M.Secure shapley value for cross-silo federated learning [J].arXiv:2209.04856,2022. [19]LUO F,AL-KUWARI S,DING Y.SVFL:Efficient Secure Aggregation and Verification for Cross-Silo Federated Learning [J].IEEE Transactions on Mobile Computing,2022(1):1-14. [20]TRAN H Y,HU J,YIN X,et al.An Efficient Privacy-Enhan-cing Cross-Silo Federated Learning and Applications for False Data Injection Attack Detection in Smart Grids [J].IEEE Transactions on Information Forensics and Security,2023,18:2538-2552. [21]LI Q,WEN Z,WU Z,et al.A Survey on Federated Learning Systems:Vision,Hype and Reality for Data Privacy and Protection [J].IEEE Transactions on Knowledge and Data Engineering,2023,35(4):3347-3366. [22]ZHANG W,ZHOU T,LU Q,et al.Dynamic-fusion-based fede-rated learning for COVID-19 detection [J].IEEE Internet of Things Journal,2021,8(21):15884-15891. [23]LI L,YU X,CAI X,et al.Contract-Theory-Based IncentiveMechanism for Federated Learning in Health CrowdSensing [J].IEEE Internet of Things Journal,2022,10(5):4475-4489. [24]LI B,SHI Y,GUO Y,et al.Incentive and Knowledge Distillation Based Federated Learning for Cross-Silo Applications [C]//IEEE INFOCOM 2022-IEEE Conference on Computer Communications Workshops.IEEE,2022:1-6. [25]TRAN V T,PHAM H H,WONG K S.Personalized privacy-preserving framework for cross-silo federated learning [J].ar-Xiv:2302.12020,2023. [26]LUO J,WU S.Adapt to Adaptation:Learning Personalization for Cross-Silo Federated Learning [C]//IJCAI:Proceedings of the Conference.2022. [27]YANG X,TAN W,PENG C,et al.Federated learning incentive mechanism design via enhanced Shapley value method [J].Wireless Communications and Mobile Computing,2022,22:1-11. [28]HUANG C,KE S,KAMHOUA C,et al.Incentivizing data contribution in cross-silo federated learning[J].arXiv:2203.03885,2022. [29]GU T,LI L,CHANG L,et al.Fair federated machine learning and its design:A comprehensive survey [J].Chinese Journal of Conputers,2020,46(9):1991-2024. [30]TRAVADI Y,PENG L,BI X,et al.Welfare and Fairness Dynamics in Federated Learning:A Client Selection Perspective [J].arXiv:2302.08976,2023. [31]SHI Z,ZHANG L,YAO Z,et al.FedFAIM:A Model Perfor-mance-based Fair Incentive Mechanism for Federated Learning [J].IEEE Transactions on Big Data,2022(1):1-13. [32]CHU L,WANG L,DONG Y,et al.Fedfair:Training fair models in cross-silo federated learning [J].arXiv:2109.05662,2021. [33]ZHANG X,DOU Y,ZHANG C,et al.Reserch on the Profit-sharing Mechanism for Federated Learning [J].Frontiers of Science and Technology of Engineering Management,2023,42(2):8-15. [34]HAN J,KHAN A F,ZAWAD S,et al.Tiff:Tokenized incentive for federated learning [C]//2022 IEEE 15th International Conference on Cloud Computing.IEEE,2022:407-416. [35]CAO X,LI Z,SUN G,et al.Cross-Silo Heterogeneous Model Federated Multitask Learning [J].arXiv:2202.08603,2022. [36]JIANG S,WU J.Coalition Formation Game in the Cross-SiloFederated Learning System [C]//2022 IEEE 19th International Conference on Mobile Ad Hoc and Smart Systems.IEEE,2022:49-57. [37]FRABONI Y,VIDAL R,LORENZI M.Free-rider attacks onmodel aggregation in federated learning [C]//International Conference on Artificial Intelligence and Statistics.PMLR,2021:1846-1854. [38]RANATHUNGA T,MCGIBNEY A,REA S,et al.Blockchain-Based Decentralized Model Aggregation for Cross-Silo Federated Learning in Industry 4.0 [J].IEEE Internet of Things Journal,2022,10(5):4449-4461. [39]WANG Y,LIG,LI K.Survey on Contribution Evaluation for Federated Learning [J].Journal of Software,2023,34(3):1168-1192. [40]YANG C,LIU J,SUN H,et al.WTDP-Shapley:Efficient and Effective Incentive Mechanism in Federated Learning for Intelligent Safety Inspection [J].IEEE Transactions on Big Data,2022(1):1-10. [41]WANG Z,YAN B,DONG A.Blockchain Empowered Federated Learning for Data Sharing Incentive Mechanism [J].Procedia Computer Science,2022,202:348-353. [42]TANG M,WONG V.An incentive mechanism for cross-silo fe-derated learning:A public goods perspective [C]//IEEE INFOCOM 2021-IEEE Conference on Computer Communications.IEEE,2021:1-10. [43]YANG Y,ZHOU Y,HU M,et al.BARA:Efficient IncentiveMechanism with Online Reward Budget Allocation in Cross-Silo Federated Learning[J].arXiv:2305.05221,2023. [44]LIU Y,TIAN M,CHEN Y,et al.A contract theory based incentive mechanism for federated learning [M]//Federated and Transfer Learning.Cham:Springer International Publishing,2022:117-137. [45]LUO X,ZHANG Z,HE J,et al.Strategic Analysis of the Pa-rameter Servers and Participants in Federated Learning:An Evolutionary Game Perspective [J].IEEE Transactions on Computational Social Systems,2024,11(1):132-143. [46]CHEN Y,ZHANG Y,WANG S,et al.Dim-ds:Dynamic incentive model for data sharing in federated learning based on smart contracts and evolutionary game theory [J].IEEE Internet of Things Journal,2022,9(23):24572-24584. [47]WANG W.Research on Incentive Mechanism of FederatedLearning Based on Repeated Game [D].Xi'an:Xidian University,2022. [48]LI Y,WANG X,ZENG R,et al.VARF:An Incentive Mechanism of Cross-silo Federated Learning in MEC [J].IEEE Internet of Things Journal,2023,10(17):15115-15132. [49]YUAN S,LIU H,LV H,et al.Adaptive incentive for cross-silo federated learning:A multi-agent reinforcement learning approach [J].arXiv:2302.07493,2023. [50]LU J,PAN B,SEID A,et al.Truthful incentive mechanism design via internalizing externalities and lp relaxation for vertical federated learning [J].IEEE Transactions on Computational Social Systems,2023,10(6):2909-2923. [51]XIONG A,CHEN Y,CHEN H,et al.A Truthful and Reliable Incentive Mechanism for Federated Learning Based on Reputation Mechanism and Reverse Auction [J].Electronics,2023,12(3):517-540. [52]XU H,NANDA P,LIANG J,et al.The Force of Compensation,a Multi-stage Incentive Mechanism Model for Federated Lear-ning[C]//International Conference on Network and System Security.Cham:Springer Nature Switzerland,2022:357-373. [53]LIN J,DU M,LIU J.Free-riders in federated learning:Attacks and defenses [J].arXiv:1911.12560,2019. [54]HUANG J,TALBI R,ZHAO Z,et al.An exploratory analysis on users' contributions in federated learning [C]//2020 Second IEEE International Conference on Trust,Privacy and Security in Intelligent Systems and Applications.IEEE,2020:20-29. [55]ZHANG N,MA Q,CHEN X.Enabling Long-Term Cooperation in Cross-Silo Federated Learning:A Repeated Game Perspective [J].IEEE Transactions on Mobile Computing,2023,22(7):3910-3924. [56]CHEN J,HU Q,JIANG H.Social welfare maximization incross-silo federated learning [C]//2022 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP 2022).IEEE,2022:4258-4262. [57]QI J,LIN F,CHEN Z,et al.High-quality model aggregation for blockchain-based federated learning via reputation-motivated task participation [J].IEEE Internet of Things Journal,2022,9(19):18378-18391. [58]PANDL K,LEISER F,THIEBES S,et al.Reward Systems for Trustworthy Medical Federated Learning [J].arXiv:2205.00470,2022. |
|