一种数据隐私保护下多方无损线性模型学习方法

doi:10.3901/JME.2023.12.017

机械工程学报 ›› 2023, Vol. 59 ›› Issue (12): 17-27.doi: 10.3901/JME.2023.12.017

• 特邀专栏：制造大数据分析与决策 • 上一篇下一篇

扫码分享

一种数据隐私保护下多方无损线性模型学习方法

华丰^1,2, 王亚森^1,2, 金骏阳³, 袁烨^1,2,3,4

1. 华中科技大学机械科学与工程学院武汉 430074;
2. 华中科技大学数字制造装备与技术国家重点实验室武汉 430074;
3. 华中科技大学无锡研究院无锡 214000;
4. 华中科技大学人工智能与自动化学院武汉 430074

收稿日期:2022-09-08 修回日期:2023-01-31 出版日期:2023-06-20 发布日期:2023-08-15
通讯作者: 袁烨(通信作者),男,1986年出生,博士,教授,博士研究生导师。主要研究方向为数据驱动建模理论及其在工业应用。E-mail:yye@hust.edu.cn
作者简介:华丰,男,1995年出生,博士研究生。主要研究方向为数据驱动机理建模在工业应用。E-mail:feng_hua@hust.edu.cn;王亚森,男,1995年出生,博士研究生。主要研究方向为机器学习,系统辨识。E-mail:arthinw@hust.edu.can;金骏阳,男,1991年出生,博士。主要研究方向为机器学习,网络拓扑结构推断。E-mail:jj415@alumni.cam.ac.uk
基金资助:
国家自然科学基金资助项目(92167201，62203182)。

A Multi-Lossless Linear Model Learning Method with Data Privacy-Preserving

HUA Feng^1,2, WANG Yasen^1,2, JIN Junyang³, YUAN Ye^1,2,3,4

1. School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074;
2. State Key Laboratory of Digital Manufacturing Equipment and Technology, Huazhong University of Science and Technology, Wuhan 430074;
3. HUST-Wuxi Research Institute, Wuxi 214000;
4. School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074

Received:2022-09-08 Revised:2023-01-31 Online:2023-06-20 Published:2023-08-15

摘要/Abstract

摘要： 随着工业大数据技术的发展，制造企业通过收集和分析生产数据，获取在预测、诊断等方面的优化方法。然而制造企业受限于建模和算力等技术瓶颈，难以高效地实现数据分析，在联合其他参与方共同协作时即需要承担信息泄露的风险，又难以保证模型的性能无损。针对这些场景问题，提出基于数据隐私保护的多方无损线性模型学习方法。首先搭建多方协作计算框架，设计了数据单向加密算法来保护数据出场的隐私安全，各协作方分别基于加密数据进行线性模型的训练。随后，研究分析线性模型与数据集的关联特性，提出线性模型无损聚合算法。最后，在典型工业场景数据集上进行方法验证，试验结果表明提出的框架可以获得性能无损的全局模型，并实现数据持有方的隐私安全保护。

关键词: 工业大数据, 隐私保护机器学习, 联邦学习, 数据分析

Abstract: With the development of industrial big data technologies, manufacturing enterprises collect and analyse production data to obtain optimization methods of forecasting and diagnosis. However, manufacturing companies are constrained by technical bottlenecks such as modelling and computing power, which make it difficult to analyse data efficiently. Manufacturing enterprises need to bear the risk of information leakage, and it is also difficult to guarantee that model performance is lossless, when cooperating with other participants. For these scenario problems, a multi-lossless linear model learning method based on data privacy preserving is proposed. Firstly, a multi-party collaborative computing framework is built and the one-way data encryption algorithm is designed to protect the privacy of the data. Each collaborator trains the linear model separately based on the encrypted data. Secondly, the study analyses the association properties of the linear model with the dataset, proposing a lossless aggregation algorithm for linear models. Finally, the method is validated on the typical industrial scenario dataset. The experimental results show that the proposed framework can obtain global models with lossless performance and achieve privacy security for the data holders.

Key words: industrial big data, privacy-preserving machine learning, federated learning, data analysis

中图分类号:

TP301

华丰, 王亚森, 金骏阳, 袁烨. 一种数据隐私保护下多方无损线性模型学习方法[J]. 机械工程学报, 2023, 59(12): 17-27.

HUA Feng, WANG Yasen, JIN Junyang, YUAN Ye. A Multi-Lossless Linear Model Learning Method with Data Privacy-Preserving[J]. Journal of Mechanical Engineering, 2023, 59(12): 17-27.

参考文献

[1] DING Han,GAO R X,ISAKSSON A J,et al. State of AI-based monitoring in smart manufacturing and introduction to focused section[J]. IEEE/ASME Transactions on Mechatronics,2020,25(5):2143-2154.
[2] YUAN Ye,MA Guijun,CHENG Cheng,et al. A general end-to-end diagnosis framework for manufacturing systems[J]. National Science Review,2020,7(2):418-429.
[3] CHENG Cheng,MA Guijun,ZHANG Yong,et al. A deep learning-based remaining useful life prediction approach for bearings[J]. IEEE/ASME Transactions on Mechatronics,2020,25(3):1243-1254.
[4] ZHAO Huimin,LIU Haodong,JIN Yang,et al. Feature extraction for data-driven remaining useful life prediction of rolling bearings[J]. IEEE Transactions on Instrumentation and Measurement,2021,70:1-10.
[5] YUAN Ye,TANG Xiuchuan,ZHOU Wei,et al. Data driven discovery of cyber physical systems[J]. Nature Communications,2019,10(1):1-9.
[6] JIN Yaochu,WANG Handing,CHUGH T,et al. Data-driven evolutionary optimization:An overview and case studies[J]. IEEE Transactions on Evolutionary Computation,2018,23(3):442-458.
[7] REN Hao,LI Hongwei,DAI Yuanshun,et al. Querying in internet of things with privacy preserving:Challenges, solutions and opportunities[J]. IEEE Network,2018,32(6):144-151.
[8] WANG Junliang,ZHENG Peng,LV Youlong,et al. Fog-IBDIS:Industrial big data integration and sharing with fog computing for manufacturing systems[J]. Engineering,2019,5(4):662-670.
[9] 李尤慧子,殷昱煜,高洪皓,等. 面向隐私保护的非聚合式数据共享综述[J]. 通信学报,2021,42(6):195-212. LI Youhuizi,YIN Yuyu,GAO Honghao,et al. Survey on privacy protection in non-aggregated data sharing[J]. Journal on Communications,2021,42(6):195-212.
[10] LI Hongwei,LIU Dongxiao,DAI Yuanshun,et al. Engineering searchable encryption of mobile cloud networks:When QoE meets QoP[J]. IEEE Wireless Communications,2015,22(4):74-80.
[11] LI Hongwei,YANG Yi,DAI Yuanshun,et al. Achieving secure and efficient dynamic searchable symmetric encryption over medical cloud data[J]. IEEE Transactions on Cloud Computing,2017,8(2):484-494.
[12] MOHASSEL P,ZHANG Yupeng. Secureml:A system for scalable privacy-preserving machine learning[C]//IEEE symposium on security and privacy,May 22-24,2017,San Jose,California. IEEE,2017:19-38.
[13] BONAWITZ K,IVANOV V,KREUTER B,et al. Practical secure aggregation for privacy-preserving machine learning[C]//Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security,October 30-November 3,2017,Dallas Texas USA. New York:ACM,2017:1175-1191.
[14] Al-RUBAIE M,CHANG J M. Privacy-preserving machine learning:Threats and solutions[J]. IEEE Security & Privacy,2019,17(2):49-58.
[15] LI Ping,LI Tong,YE Heng,et al. Privacy-preserving machine learning with multiple data providers[J]. Future Generation Computer Systems,2018,87:341-350.
[16] MCMAHAN B,MOORE E,RAMAGE D,et al. Communication-efficient learning of deep networks from decentralized data[C]//Artificial Intelligence and Statistics,20-22 April 2017,Fort Lauderdale,USA. JMLR:W&CP volume 54,2017:1273-1282.
[17] BONAWITZ K,EICHNER H,GRIESKAMP W,et al. Towards federated learning at scale:System design[J]. Proceedings of Machine Learning and Systems,2019,1:374-388.
[18] KONECNY J,MCMAHAN H B,YU F X,et al. Federated learning:Strategies for improving communication efficiency[J]. arXiv preprint arXiv:1610.05492,2016.
[19] YANG Qiang,LIU Yang,CHENG Yong,et al. Federated learning[C]//Synthesis Lectures on Artificial Intelligence and Machine Learning, Switzerland, 2019.
[20] YANG Qiang,LIU Yang,CHEN Tianjian,et al. Federated machine learning:Concept and applications[J]. ACM Transactions on Intelligent Systems and Technology,2019,10(2):1-19.
[21] RIVEST R L,ADLEMAN L,DERTOUZOS M L. On data banks and privacy homomorphisms[J]. Foundations of secure computation,1978,4(11):169-180.
[22] YI Xun,PAULET R,BERTINO E. Homomorphic encryption[M]//Homomorphic encryption and applications. Springer,Cham,2014:27-46.
[23] FAN Junfeng,VERCAUTEREN F. Somewhat practical fully homomorphic encryption[J]. Cryptology ePrint Archive,2012.
[24] BRAKERSKI Z,GENTRY C,VAIKUNTANATHAN V. (Leveled) fully homomorphic encryption without bootstrapping[J]. ACM Transactions on Computation Theory,2014,6(3):1-36.
[25] ACAR A,AKSU H,ULUAGAC A S,et al. A survey on homomorphic encryption schemes:Theory and implementation[J]. ACM Computing Surveys,2018,51(4):1-35.
[26] DWORK C,MCSHERRY F,NISSIM K,et al. Calibrating noise to sensitivity in private data analysis[C]//Theory of cryptography conference,March 4-72006,New York,USA. Berlin,Heidelberg:Springer,2006:265-284.
[27] DWORK C,ROTH A. The algorithmic foundations of differential privacy[J]. Foundations and Trends in Theoretical Computer Science,2014,9(3-4):211-407.
[28] ABAD M,CHU A,GOODFELLOW I,et al. Deep learning with differential privacy[C]//Proceedings of the ACM SIGSAC conference on computer and communications security,October 24-282016,Vienna,Austria. New York:ACM,2016:308-318.
[29] ZHANG Jun,ZHANG Zhenjie,XIAO Xiaokui,et al. Functional mechanism:regression analysis under differential privacy[J]. arXiv preprint arXiv:1208.0219, 2012.

一种数据隐私保护下多方无损线性模型学习方法

A Multi-Lossless Linear Model Learning Method with Data Privacy-Preserving

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

本文评价

[1]	李响, 付春霖, 雷亚国, 李乃鹏, 杨彬. 保证数据隐私的装备协同智能故障诊断联邦迁移学习方法[J]. 机械工程学报, 2023, 59(6): 1-9.
[2]	邵海东, 肖一鸣, 闵志闪, 韩淞宇, 张海舟. 区块链和边缘计算赋能的联邦学习故障诊断框架[J]. 机械工程学报, 2023, 59(21): 283-292.
[3]	王磊, 金校, 唐红涛, 李西兴, 李益兵, 郭顺生, 官思佳. 基于联邦学习框架的制造服务个性化推荐方法研究[J]. 机械工程学报, 2023, 59(12): 149-161.
[4]	汪俊亮, 高鹏捷, 张洁, 王力翚. 制造大数据分析综述：内涵、方法、应用和趋势[J]. 机械工程学报, 2023, 59(12): 1-16.
[5]	蒋仁言. 两类高度截尾数据及其参数估计问题[J]. 机械工程学报, 2023, 59(10): 374-382.
[6]	姜生元, 梁杰能, 赖小明, 邓湘金, 庞勇, 张伟伟, 唐钧跃, 全齐全, 彭兢, 张高, 邓宗全. 嫦娥五号月壤剖面钻进取芯状态分析与解译[J]. 机械工程学报, 2022, 58(10): 348-360.
[7]	方伟光, 郭宇, 黄少华, 刘道元, 崔世婷, 廖文和, 洪东跑. 大数据驱动的离散制造车间生产过程智能管控方法研究[J]. 机械工程学报, 2021, 57(20): 277-291.
[8]	李宁,余进. 列车能耗测试装置与分析方法研究^*[J]. 电气工程学报, 2020, 15(3): 50-56.
[9]	刘阳, 郜志英, 周晓敏, 张清东. 工业数据驱动下薄板冷轧颤振的LSTM智能预报[J]. 机械工程学报, 2020, 56(11): 121-131.
[10]	佘承其, 张照生, 刘鹏, 孙逢春. 大数据分析技术在新能源汽车行业的应用综述——基于新能源汽车运行大数据[J]. 机械工程学报, 2019, 55(20): 3-16.
[11]	黎敏, 谢玄, 陈泽, 杨孟瑶, 杨德斌, 蒋靖. 基于函数型数据分析的半导体生产过程监控[J]. 机械工程学报, 2018, 54(16): 62-69.
[12]	雷亚国, 贾峰, 周昕, 林京. 基于深度学习理论的机械装备大数据健康监测方法[J]. 机械工程学报, 2015, 51(21): 49-56.
[13]	倪敬;项占琴;潘晓弘;吕福在. 管捆成形电液系统自学习粗糙－模糊PID控制研究[J]. , 2006, 42(10): 224-228.