高级搜索

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于混合隐私的区块链高效模型协同训练共享方案

张翠 杨辉 王寒凝 王江 曾创展 李荣宽

张翠, 杨辉, 王寒凝, 王江, 曾创展, 李荣宽. 基于混合隐私的区块链高效模型协同训练共享方案[J]. 电子与信息学报, 2023, 45(3): 775-783. doi: 10.11999/JEIT221104
引用本文: 张翠, 杨辉, 王寒凝, 王江, 曾创展, 李荣宽. 基于混合隐私的区块链高效模型协同训练共享方案[J]. 电子与信息学报, 2023, 45(3): 775-783. doi: 10.11999/JEIT221104
ZHANG Cui, YANG Hui, WANG Hanning, WANG Jiang, ZENG Chuangzhan, LI Rongkuan. Efficient Model Collaborative Training and Sharing Scheme of Blockchain Based on Hybrid Privacy[J]. Journal of Electronics & Information Technology, 2023, 45(3): 775-783. doi: 10.11999/JEIT221104
Citation: ZHANG Cui, YANG Hui, WANG Hanning, WANG Jiang, ZENG Chuangzhan, LI Rongkuan. Efficient Model Collaborative Training and Sharing Scheme of Blockchain Based on Hybrid Privacy[J]. Journal of Electronics & Information Technology, 2023, 45(3): 775-783. doi: 10.11999/JEIT221104

基于混合隐私的区块链高效模型协同训练共享方案

doi: 10.11999/JEIT221104
基金项目: 国家自然科学基金(62122015)
详细信息
    作者简介:

    张翠:女,博士生,研究方向为区块链技术

    杨辉:男,教授,研究方向为光通信与光网络、6G光与无线融合、人工智能网络、区块链等

    王寒凝:女,高级工程师,研究方向为数据管理、数据分析

    王江:男,博士,研究方向为音视频分析处理、隐私计算

    曾创展:男,博士,研究方向为人工智能、数据分析

    李荣宽:男,高级工程师,研究方向为信息服务、数据服务等

    通讯作者:

    杨辉 yanghui@bupt.edu.cn

  • 中图分类号: TN929.5; TP311.13

Efficient Model Collaborative Training and Sharing Scheme of Blockchain Based on Hybrid Privacy

Funds: The National Natural Science Foundation of China (62122015)
  • 摘要: 针对海量数据下,基于区块链的联邦学习数据共享平台面临的效率低下和隐私泄露问题,该文提出基于混合隐私的区块链高效模型协同训练共享方案。在该方案中,首先根据欧氏距离设计了一种基于相似度的训练成员选择算法来选择训练成员,组成联邦社区,即通过选取少量的高匹配训练节点来提高训练的效率和效果。然后,结合阈值同态加密和差分隐私,设计一种基于混合隐私技术的模型协同训练共享方案来保证训练和共享过程中的隐私性。实验结果和系统实现表明,所提方案可以在保证训练结果准确率的情况下,实现高效训练和隐私保护下的数据共享。
  • 图  1  高效模型协同训练共享模型架构图

    图  2  共享模型协同训练流程图

    图  3  智能合约的主要gas消耗

    图  4  训练成员数目对基于相似度的训练成员选择算法的效果对比图

    图  5  训练成员数据量对基于相似度的训练成员选择算法的效果对比图

    图  6  基于相似度的训练成员选择算法效果图

    图  7  聚合不同数量噪声对准确度的影响

    图  8  数据信息浏览界面

    图  9  联邦学习训练模块展示图

    算法1 基于相似度的训练成员选择算法
     输入:数据共享请求$R = \{ Q,C{{ = (} }C_{{1} }^q,C_2^q,\cdots,C_{ {n} }^q{{)} }\}$,所有数据
        拥有者$ P $的对应摘要集合$A{\mathbf{ = (} }{A_{{1} } },{A_{{2} } },\cdots,{A_N}{{)} }$,其中
        ${A_i}{{ = (} }{C_{i{{1} } } },{C_{i2} },\cdots,{C_{in} }{{)} }$;联邦社区成员个数$ h $
     输出:联邦学习社区成员$F{{ = (} }{P_{{1} } },{P_{{2} } },\cdots,{P_h}{{)} }$
     初始化$F{{[\;] = \{ 0\} } }$, $r = s = 0$
     for $ i = 1;i < N + 1;i + + $ do
       for $j = 1;j < n + 1;j + + $ do
         $ r = {(C_1^q - {C_{ij}})^2} $
         $ s = s + r $
       end for
         $ S({P^q},{P_i}) = \sqrt s $
         $S({P^q},{P_i}) \to F{{[\;]} }$
     end for
         ${\text{Rank} }(F{{[\;]} })$
       从$ F{\mathbf{[]}} $中选取前$ h $个数据拥有者组成$F{\mathbf{ = (} }{P_{{1} } },{P_{{2} } },\cdots,{P_h}{{)} }$
     return $F{\mathbf{ = (} }{P_{{1} } },{P_{{2} } },\cdots,{P_h}{{)} }$
    下载: 导出CSV
  • [1] KASUBI J W, MANJAIAH D H, and DEMEWEZ G D. A review on the internet of things and big data analytics based on smart cities[C]. 2021 International Conference on Artificial Intelligence and Smart Systems (ICAIS), Coimbatore, India, 2021: 1352–1358.
    [2] LV Zhihan, LOU Ranran, LI Jinhua, et al. Big data analytics for 6G-enabled massive internet of things[J]. IEEE Internet of Things Journal, 2021, 8(7): 5350–5359. doi: 10.1109/JIOT.2021.3056128
    [3] DINH T N, SHEN Yilin, and THAI M T. The walls have ears: Optimize sharing for visibility and privacy in online social networks[C]. The 21st ACM International Conference on Information and Knowledge Management, Hawaii, USA, 2012: 1452–1461.
    [4] LIU Chi, ZHU Tianqing, ZHANG Jun, et al. Privacy intelligence: A survey on image sharing on online social networks[J]. arXiv preprint arXiv: 2008.12199, 2020.
    [5] ALEDHARI M, RAZZAK R, PARIZI R M, et al. Federated learning: A survey on enabling technologies, protocols, and applications[J]. IEEE Access, 2020, 8: 140699–140725. doi: 10.1109/ACCESS.2020.3013541
    [6] YANG Hui, YUAN Jiaqi, LI Chao, et al. BrainIoT: Brain-like productive services provisioning with federated learning in industrial IoT[J]. IEEE Internet of Things Journal, 2022, 9(3): 2014–2024. doi: 10.1109/JIOT.2021.3089334
    [7] YANG Hui, LIANG Yongshen, YUAN Jiaqi, et al. Distributed blockchain-based trusted multidomain collaboration for mobile edge computing in 5G and beyond[J]. IEEE Transactions on Industrial Informatics, 2020, 16(11): 7094–7104. doi: 10.1109/TII.2020.2964563
    [8] YANG Hui, YUAN Jiaqi, YAO Haipeng, et al. Blockchain-based hierarchical trust networking for JointCloud[J]. IEEE Internet of Things Journal, 2020, 7(3): 1667–1677. doi: 10.1109/JIOT.2019.2961187
    [9] LI Chao, YANG Hui, BAO Bowen, et al. A secure device access based on Blockchain for IoT in smart city[C]. 2021 International Wireless Communications and Mobile Computing (IWCMC), Harbin City, China, 2021: 1172–1174.
    [10] YANG Hui, BAO Bowen, LI Chao, et al. Blockchain-enabled tripartite anonymous identification trusted service provisioning in industrial IoT[J]. IEEE Internet of Things Journal, 2022, 9(3): 2419–2431. doi: 10.1109/JIOT.2021.3097440
    [11] LI Chao, YANG Hui, SUN Zhengjie, et al. Federated hierarchical trust-based interaction scheme for cross-domain industrial IoT[J/OL]. IEEE Internet of Things Journal, 2022, https://ieeexplore.ieee.org/document/9865221, 2022.
    [12] ZHANG Xu and HOU Haibo. Application of federated learning in industrial internet with device identifier[C]. 2021 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB), Chengdu, China, 2021: 1–5
    [13] 陈乃月, 金一, 李浥东, 等. 基于区块链的公平性联邦学习模型[J]. 计算机工程, 2022, 48(6): 33–41. doi: 10.19678/j.issn.1000-3428.0064095

    CHEN Naiyue, JIN Yi, LI Yidong, et al. Federated learning model with fairness based on Blockchain[J]. Computer Engineering, 2022, 48(6): 33–41. doi: 10.19678/j.issn.1000-3428.0064095
    [14] WENG Jiasi, WENG Jian, ZHANG Jilian, et al. DeepChain: Auditable and privacy-preserving deep learning with Blockchain-based incentive[J]. IEEE Transactions on Dependable and Secure Computing, 2021, 18(5): 2438–2455. doi: 10.1109/TDSC.2019.2952332
    [15] QI Yuanhang, HOSSAIN M S, NIE Jiangtian, et al. Privacy-preserving blockchain-based federated learning for traffic flow prediction[J]. Future Generation Computer Systems, 2021, 117: 328–337. doi: 10.1016/j.future.2020.12.003
    [16] 汤凌韬, 王迪, 张鲁飞, 等. 基于安全多方计算和差分隐私的联邦学习方案[J]. 计算机科学, 2022, 49(9): 297–305. doi: 10.11896/jsjkx.210800108

    TANG Lingtao, WANG Di, ZHANG Lufei, et al. Federated learning scheme based on secure multi-party computation and differential privacy[J]. Computer Science, 2022, 49(9): 297–305. doi: 10.11896/jsjkx.210800108
    [17] TRUEX S, BARACALDO N, ANWAR A, et al. A hybrid approach to privacy-preserving federated learning[C]. Proceedings of the 12th ACM Workshop on Artificial Intelligence and Security, London, UK, 2019: 1–11.
    [18] SHE Wei, LIU Qi, TIAN Zhao, et al. Blockchain trust model for malicious node detection in wireless sensor networks[J]. IEEE Access, 2019, 7: 38947–38956. doi: 10.1109/ACCESS.2019.2902811
    [19] 柴迪. 阈值同态加密在隐私计算中的应用[J]. 信息通信技术与政策, 2021, 47(7): 82–86. doi: 10.12267/j.issn.2096-5931.2021.07.012

    CHAI Di. Application of threshold homomorphic encryption in privacy computing[J]. Information and Communications Technology and Policy, 2021, 47(7): 82–86. doi: 10.12267/j.issn.2096-5931.2021.07.012
    [20] GENG Ziye, HE Yunhua, WANG Chao, et al. A Blockchain based privacy-preserving reputation scheme for cloud service[C]. ICC 2021 - IEEE International Conference on Communications, Montreal, Canada, 2021: 1–6.
  • 加载中
图(9) / 表(1)
计量
  • 文章访问数:  583
  • HTML全文浏览量:  273
  • PDF下载量:  132
  • 被引次数: 0
出版历程
  • 收稿日期:  2022-08-23
  • 修回日期:  2022-12-30
  • 网络出版日期:  2023-01-05
  • 刊出日期:  2023-03-10

目录

    /

    返回文章
    返回