一种抗分布式机器学习恶意节点的区块链方案

doi:10.19665/j.issn1001-2400.2023.02.018

摘要/Abstract

摘要：

现有的分布式学习方案大多通过在协议中添加惩戒机制来解决恶意节点问题。此类方法基于两个假设:一是参与方为自身利益最大化而放弃恶意行为的假设,在事件发生后才可以对计算结果进行验证,不适用于一些需要即时验证的场景;二是基于可信第三方的假设,然而在实际中第三方的可信度却无法完全保证。利用区块链的信任机制,针对该问题提出一种基于区块链的抗恶意节点方案——将机器学习中模型训练的全过程通过智能合约实现,以确保机器学习过程不被恶意节点破坏。本方案以基于安全多方计算的分布式机器学习模型为研究模型,利用区块链的智能合约来实现数据的共享、验证和训练过程,所有参与方均只能按照指定的协议执行,将所有参与方转换为半诚实参与方;同时,为解决区块链公开透明特性带来的隐私问题,利用环签名隐藏参与方的数据地址,保护参与方的身份。与传统基于安全多方计算的分布式机器学习模型进行比较,表明本方案在抵抗恶意节点方面具有较强的优越性。

关键词: 区块链, 智能合约, 分布式学习, 恶意节点, 安全多方计算, 环签名

Abstract:

Most of the existing distributed learning schemes solve the problem of malicious nodes by adding a disciplinary mechanism to the protocol.This method is based on two assumptions:1.Participants give up the assumption of malicious behavior to maximize their own interests,and the calculation results can be verified only after the event occurs,which is not suitable for some scenarios requiring immediate verification;2.It is based on the assumption of a trusted third party.However,in practice,the credibility of the third party cannot be fully guaranteed.Using the trust mechanism of the blockchain,this paper proposes an anti malicious node scheme based on the smart contract,which realizes the whole process of model training in machine learning through the smart contract to ensure that the machine learning model is not damaged by malicious nodes.This scheme takes the distributed machine learning model based on secure multi-party computing as the research model,and uses the smart contract of the blockchain to realize the data sharing,verification and training process.All participants can only execute according to the specified protocol,converting all participants into semi sincere participants;At the same time,in order to solve the privacy problems brought by the open and transparent characteristics of the blockchain,ring signature is used to hide the data address of participants and protect the identity of participants.Experiments show that this scheme has great advantages in resisting malicious nodes compared with the traditional distributed machine learning model based on secure multi-party computing.

Key words: blockchain, smart contract, distributed learning, malicious nodes,secure multi-party computation, ring signature

中图分类号:

TP311.1

刘远振, 杨颜博, 张嘉伟, 李宝山, 马建峰. 一种抗分布式机器学习恶意节点的区块链方案[J]. 西安电子科技大学学报, 2023, 50(2): 178-187.

LIU Yuanzhen, YANG Yanbo, ZHANG Jiawei, LI Baoshan, MA Jianfeng. Blockchain scheme for anti malicious nodes in distributed machine learning[J]. Journal of Xidian University, 2023, 50(2): 178-187.

图/表 5

参考文献 26

[1]	高德勇, 康自兵, 王松, 等. 利用卷积块注意力机制识别人体动作的方法[J]. 西安电子科技大学学报, 2022, 49(4):144-155.
	GAO Deyong, KANG Zibing, WANG Song, et al. Method to Recognize Human Action by Using the Convolutional Block Attention Mechanism[J]. Journal of Xidian University, 2022, 49(4):144-155.
[2]	胡代旺, 焦一源, 李雁妮. 一种新型高效的文库知识图谱实体关系抽取算法[J]. 西安电子科技大学学报, 2021, 48(6):75-83.
	HU Daiwang, JIAO Yiyuan, LI Yanni. Novel and Efficient Algorithm for Entity Relation Extraction with the Corpus Knowledge Graph[J]. Journal of Xidian University, 2021, 48(6):75-83.
[3]	胡奥婷, 胡爱群, 胡韵, 等. 机器学习中差分隐私的数据共享及发布:技术、应用和挑战[J]. 信息安全学报, 2022, 7(4):1-16.
	HU Aoting, HU Aiqun, HU Yun, et al. Differentially Private Data Sharing and Publishing in Machine Learning:Techniques,Applications,and Challenges[J]. Journal of Cyber Security, 2022, 7(4):1-16.
[4]	王腾, 霍峥, 黄亚鑫, 等. 联邦学习中的隐私保护技术研究综述[J]. 计算机应用, 2023, 43(2):437-449. doi: 10.11772/j.issn.1001-9081.2021122072
	WANG Teng, HUO Zheng, HUANG Yaxin, et al. Review on Privacy-Preserving Technologies in Federated Learning[J]. Journal of Computer Applications, 2023, 43(2):437-449. doi: 10.11772/j.issn.1001-9081.2021122072
[5]	莫康华. 面向机器学习隐私安全的对抗攻击技术研究[D]. 广州: 广州大学, 2022.
[6]	ZHANG D, YU F R, YANG R. Blockchain-Based Distributed Software-Defined Vehicular Networks:A Dueling Deep Q-Learning Approach[J/OL].[2022-06-10].DOI:10.1109/TCCN.2019.2944399. doi: 10.1109/TCCN.2019.2944399
[7]	孙文礼. 基于安全多方计算的隐私保护支持向量机算法研究[D]. 哈尔滨: 哈尔滨工业大学, 2019.
[8]	汤漾. 基于安全多方计算的聚类算法统一框架的设计与实现[D]. 北京: 北京邮电大学, 2019.
[9]	梁伦. 基于区块链的拜占庭容错分布式机器学习算法研究[D]. 南京: 东南大学, 2021.
[10]	XIAO L, DING Y, JIANG D, et al. A Reinforcement Learning and Blockchain-Based Trust Mechanism for Edge Networks[J/OL].[2022-06-12].DOI:10.1109/TCOMM.2020.2995371. doi: 10.1109/TCOMM.2020.2995371
[11]	HUA G, ZHU L, WU J, et al. Blockchain-Based Federated Learning for Intelligent Control in Heavy Haul Railway[J/OL].[2022-06-11].DOI:10.1109/ACCESS.2020.3021253. doi: 10.1109/ACCESS.2020.3021253
[12]	黄建华, 江亚慧, 李忠诚. 利用区块链构建公平的安全多方计算[J]. 计算机应用研究, 2020, 37(1):225-230.
	HUANG Jianhua, JIANG Yahui, LI Zhongcheng. Constructing Fai Secure Multi-Party Computation Based on Blockchain[J]. Application Research of Computers, 2020, 37(1):225-230.
[13]	郭娟娟, 王琼霄, 许新. 安全多方计算及其在机器学习中的应用[J]. 计算机研究与发展, 2021, 58(10):2163-2186.
	GUO Juanjuan, WANG Qiongxiao, XU Xin. Secure Multiparty Computation and Application in Machine Learning[J]. Journal of Computer Research and Development, 2021, 58(10):2163-2186.
[14]	DENG L, JIANG Y, NING B. Identity-Based Linkable Ring Signature Scheme[J]. IEEE ACCESS, 2019, 7:153969-153976. doi: 10.1109/ACCESS.2019.2948972
[15]	陈思吉, 翟社平, 汪一景. 一种基于环签名的区块链隐私保护算法[J]. 西安电子科技大学学报, 2020, 47(5):86-93.
	CHEN Siji, ZHAI Sheping, WANG Yijing. Blockchain Privacy Protection Algorithm Based on Ring Signature[J]. Journal of Xidian University, 2020, 47(5):86-93.
[16]	FC A, TKD B, CP A. A Systematic Literature Review of Blockchain-Based Applications:Current Status,Classification and Open Issues - Science Direct[J]. Telematics and Informatics, 2019, 36:55-81. doi: 10.1016/j.tele.2018.11.006
[17]	ZHENG Z, XIE S, DAI H, et al. An Overview of Blockchain Technology:Architecture,Consensus,and Future Trends[C]// 2017 IEEE 6th International Congress on Big Data(BIGDATA CONGRESS 2017).Piscataway:IEEE, 2017:557-564.
[18]	王慧, 王励成, 柏雪, 等. 区块链隐私保护和扩容关键技术研究[J]. 西安电子科技大学学报, 2020, 47(5):28-39.
	WANG Hui, WANG Licheng, BAI Xue, et al. Research on Key Technology of Blockchain Privacy Protection and Scalability[J]. Journal of Xidian University, 2020, 47(5):28-39.
[19]	QIU C, YU F R, YAO H, et al. Blockchain-Based Software-Defined Industrial Internet of Things:A Dueling Deep Q-Learning Approach[J/OL].[2022-06-08].DOI:10.1109/JIOT.2018.2871394. doi: 10.1109/JIOT.2018.2871394
[20]	ZERKA F, UROVI V, VAIDYANATHAN A, et al. Blockchain for Privacy Preserving and Trustworthy Distributed Machine Learning in Multicentric Medical Imaging (C-DistriM)[J]. IEEE Access, 2020, 8:183939-183951. doi: 10.1109/Access.6287639
[21]	杨颜博, 张嘉伟, 马建峰. 一种使用区块链保护车联网数据隐私的方法[J]. 西安电子科技大学学报, 2021, 48(3):21-30.
	YANG Yanbo, ZHANG Jiawei, MA Jianfeng. Method for Using the Blockchain to Protect Data Privacy of IoV[J]. Journal of Xidian University, 2021, 48(3):21-30.
[22]	PARK S. A Distributed Oracle Using Intel SGX for Blockchain-Based IoT Applications[J/OL].[2022-06-15].DOI:10.3390/s20092725. doi: 10.3390/s20092725
[23]	WOO S, SONG J, PARK S. A Distributed Oracle Using Intel SGX for Blockchain-Based IoT Applications[J/OL].[2020-06-11].DOI:10.3390/s20092725. doi: 10.3390/s20092725
[24]	ALBREIKI H, REHMAN M, SALAH K, et al. Trustworthy Blockchain Oracles:Review,Comparison,and Open Research Challenges[J]. IEEE Access, 2020, 8:85675-85685. doi: 10.1109/Access.6287639
[25]	王影军. 基于区块链与云-边缘计算的车联网数据安全存储与共享研究[D]. 赣州: 江西理工大学, 2022.
[26]	罗金喜, 王珺. 边缘计算中基于区块链的访问控制机制[J]. 计算机时代, 2022(7):12-16.
	LUO Jinxi, WANG Jun. Blockchain-based Access Control Mechanism in Edge Computing[J]. Computer Era, 2022(7):12-16.

运行阶段	数据共享	数据验证	数据训练	模型聚合
SMC	65 856	5	7 767	5
SCMPC4	12 915	135	10 311	69
SCMPC8	28 689	125	8 275	92
SCMPC12	36 670	123	9 821	99