机器学习越来越多地被用于解决工程问题,其性能通过准确性、效率和安全性来衡量。值得注意的是,当安全性成为需要特别关注的问题时,区块链技术被引入机器学习中。然而,现有解决方案存在一个研究空白,即主要关注使用区块链保护数据安全,而忽略了计算安全性,这使得传统的机器学习过程容易受到链下风险的影响。因此,本研究旨在开发一个名为机器学习在区块链上(machine learning on blockchain, MLOB)的框架,以确保数据和计算过程的安全性。其核心原则是将数据和计算过程都置于区块链上,作为区块链智能合约执行,并保护链上的执行记录。通过原型开发并使用工业检查的案例研究进一步校准,建立了该框架。结果表明,与现有的机器学习和区块链隔离解决方案相比,MLOB框架在安全性方面更优越(成功防御了六种设计的攻击场景),保持了准确性(与基线相比差异为0.01%),尽管效率略有下降(延迟增加了0.231 s)。关键发现表明,MLOB可以显著增强工程计算的计算安全性,而不增加计算能力需求。这一发现可以缓解对机器学习与区块链集成所需计算资源的担忧。通过适当调整,MLOB框架可以为更广泛的工程挑战提供各种新颖的解决方案,以实现计算安全性。
Abstract
Machine learning (ML) has been increasingly adopted to solve engineering problems with performance gauged by accuracy, efficiency, and security. Notably, blockchain technology (BT) has been added to ML when security is a particular concern. Nevertheless, there is a research gap that prevailing solutions focus primarily on data security using blockchain but ignore computational security, making the traditional ML process vulnerable to off-chain risks. Therefore, the research objective is to develop a novel ML on blockchain (MLOB) framework to ensure both the data and computational process security. The central tenet is to place them both on the blockchain, execute them as blockchain smart contracts, and protect the execution records on-chain. The framework is established by developing a prototype and further calibrated using a case study of industrial inspection. It is shown that the MLOB framework, compared with existing ML and BT isolated solutions, is superior in terms of security (successfully defending against corruption on six designed attack scenario), maintaining accuracy (0.01% difference with baseline), albeit with a slightly compromised efficiency (0.231 second latency increased). The key finding is MLOB can significantly enhances the computational security of engineering computing without increasing computing power demands. This finding can alleviate concerns regarding the computational resource requirements of ML–BT integration. With proper adaption, the MLOB framework can inform various novel solutions to achieve computational security in broader engineering challenges.
本研究致力于构建一个可部署的框架,结合机器学习和区块链技术,以增强工程中的计算安全性。这种机器学习在区块链上(machine learning on blockchain, MLOB)框架的基本原则是将所有工程计算记录和交易(包括数据、计算过程和结果)以最少或无人为干预的方式嵌入区块链中。为了说明MLOB框架,基于一个实际案例开发了一个原型。虽然这种框架增强了计算安全性,但可能会在效率和准确性方面产生成本。这种现象可以称为准确性、效率和安全性的“平衡三角形”,意味着在评估计算过程的性能时,需要平衡这三个指标。
机器学习与区块链的集成可以是以区块链为中心或以机器学习为中心[35]。前者涉及使用机器学习分析存储在区块链上的数据以获得有意义的见解。例如,Shinde等[36]提出了一种增强的道路建设管理流程,其中机器学习算法用于预测,区块链用于确保利益相关者之间合同的安全性。Wong等[37]探讨了机器学习在稀缺资源调度和管理中的应用,区块链用于保护隐私数据免受网络威胁。Xu等[38]开发了一个基于区块链的框架,名为P-OSH,强调在建筑职业安全与健康(occupational safety and health, OSH)管理中的隐私保护。Kallapu等[39]提出了一种基于区块链的属性感知加密方法,用于安全通信,并在最新研究中提出了一种安全传输图像的新方法[40]。
BuedeDM, MillerWD. The engineering design of systems: models and methods. Hoboken: John Wiley & Sons; 2016.
[2]
SinhaR, ParedisCJ, LiangVC, KhoslaPK. Modeling and simulation methods for design of engineering systems. J Comput Inf Sci Eng2001;1(1):84‒91. . 10.1115/1.1344877
[3]
NagrathI, GopalM. Control systems engineering. Hong Kong: New Age International; 2006.
[4]
SioshansiR, ConejoAJ. Optimization in engineering. Cham: Springer International Publishing; 2017. . 10.1007/978-3-319-56769-3
[5]
WillardJ, JiaX, XuS, SteinbachM, KumarV. Integrating scientific knowledge with machine learning for engineering and environmental systems. ACM Comput Surv2023;55(4):1‒37. . 10.1145/3514228
[6]
PenumuruDP, MuthuswamyS, KarumbuP. Identification and classification of materials using machine vision and machine learning in the context of Industry 4.0. J Intell Manuf2020;31(5):1229‒41. . 10.1007/s10845-019-01508-6
[7]
FrankM, DrikakisD, CharissisV. Machine-learning methods for computational science and engineering. Computation2020;8(1):15. . 10.3390/computation8010015
[8]
MasanaM, LiuX, TwardowskiB, MentaM, BagdanovAD, Van De WeijerJ. Class-incremental learning: survey and performance evaluation on image classification. IEEE Trans Pattern Anal Mach Intell2023;45(5):5513‒33. . 10.1109/tpami.2022.3213473
[9]
JeongD. Artificial intelligence security threat, crime, and forensics: taxonomy and open issues. IEEE Access2020;8:184560‒74. . 10.1109/access.2020.3029280
[10]
NadeemA, VosD, CaoC, PajolaL, DieckS, BaumgartnerR, et al. SoK: explainable machine learning for computer security applications. In: Proceedings of the 2023 IEEE 8th European Symposium on Security and Privacy (EuroS&P); 2023 Jul 3‒7; Delft, Netherlands. New York City: IEEE; 2023. p. 221‒40. . 10.1109/eurosp57164.2023.00022
[11]
QadirS, NoorB. Applications of machine learning in digital forensics. In: Proceedings of the 2021 International Conference on Digital Futures and Transformative Technologies (ICoDT2); 2021 May 20‒21; Islamabad, Pakistan. New York City: IEEE; 2021. p. 1‒8. . 10.1109/icodt252288.2021.9441543
[12]
NewhartKB, MarksCA, Rauch-WilliamsT, CathTY, HeringAS. Hybrid statistical-machine learning ammonia forecasting in continuous activated sludge treatment for improved process control. J Water Process Eng2020;37:101389. . 10.1016/j.jwpe.2020.101389
[13]
LuW, LouJ, WuL. Combining smart construction objects-enabled blockchain oracles and signature techniques to ensure information authentication and integrity in construction. J Comput Civ Eng2023;37(6):04023031. . 10.1061/jccee5.cpeng-5268
[14]
VenkatesanC, KarthigaikumarP, VaratharajanR. A novel LMS algorithm for ECG signal preprocessing and KNN classifier based abnormality detection. Multimedia Tools Appl2018;77(8):10365‒74. . 10.1007/s11042-018-5762-6
[15]
LuW, WuL, ZhaoR, LiX, XueF. Blockchain technology for governmental supervision of construction work: learning from digital currency electronic payment systems. J Constr Eng Manage2021;147(10):04021122. . 10.1061/(asce)co.1943-7862.0002148
[16]
LuW, WuL, ZhaoR. Rebuilding trust in the construction industry: a blockchain-based deployment framework. Int J Constr Manag2023;23(8):1405‒16. . 10.1080/15623599.2021.1974683
[17]
PapernotN, McDanielP, SinhaA, WellmanMP. SoK: security and privacy in machine learning. In: Proceedings of the 2018 IEEEEuropean Symposium on Security and Privacy EuroS&P); 2018 Apr 24‒26; London, UK. New York City: IEEE; 2018. p. 399‒414. . 10.1109/eurosp.2018.00035
[18]
CurrieR. Software engineer accused of stealing $300k from employer was ‘inspired by office space’ [Internet]. San Francisco: Situation Publishing; 2023 Jan 13 [cited 2023 Feb 6]. Available from:
[19]
FangM, CaoX, JiaJ, GongNZ. Local model poisoning attacks to byzantine-robust federated learning. In: Proceedings of the 29th USENIX Conference on Security Symposium; 2020 Aug 12‒14; Berkeley, CA, USA. Berkeley: The USENIX Association; 2020. p. 1623‒40.
[20]
LeeJH, ShinJ, RealffMJ. Machine learning: overview of the recent progresses and implications for the process systems engineering field. Comput Chem Eng2018;114:111‒21. . 10.1016/j.compchemeng.2017.10.008
[21]
GuoJ, GaoH, LiuZ, HuangF, ZhangJ, LiX, et al. ICRA: an intelligent clustering routing approach for UAV ad hoc networks. IEEE Trans Intell Transp Syst2023;24(2):2447‒60. . 10.1109/tits.2022.3145857
[22]
YuanL, LuW, XueF, LiM. Building feature-based machine learning regression to quantify urban material stocks: a Hong Kong study. J Ind Ecol2023;27(1):336‒49. . 10.1111/jiec.13348
[23]
YuanL, LuW, WuY. Characterizing the spatiotemporal evolution of building material stock in China’s greater bay area: a statistical regression method. J Ind Ecol2023;27(6):1553‒66. . 10.1111/jiec.13438
[24]
GuoJ, LiX, LiuZ, MaJ, YangC, ZhangJ, et al. Trove: a context-awareness trust model for vanets using reinforcement learning. IEEE Internet Things J2020;7(7):6647‒62. . 10.1109/jiot.2020.2975084
[25]
DongZ, ChenJ, LuW. Computer vision to recognize construction waste compositions: a novel boundary-aware transformer (BAT) model. J Environ Manage2022;305:114405. . 10.1016/j.jenvman.2021.114405
ParasharA, ParasharA, ShabazM, GuptaD, SahuAK, KhanMA. Advancements in artificial intelligence for biometrics: a deep dive into model-based gait recognition techniques. Eng Appl Artif Intell2024;130:107712. . 10.1016/j.engappai.2023.107712
[28]
AsmithaP, RupaC, NikithaS, HemalathaJ, SahuAK. Improved multi-view biometric object detection for anti spoofing frauds. Multimedia Tools Appl2024;83(33):80161. . 10.1007/s11042-024-18458-8
[29]
TeohYK, GillSS, ParlikadAK. Iot and fog computing based predictive maintenance model for effective asset management in Industry 4.0 using machine learning. IEEE Internet Things J2021;10(3):2087‒94.
[30]
LăzăroiuG, AndronieM, IataganM, GeamănuM, ȘtefănescuR, DijmărescuI. Deep learning-assisted smart process planning, robotic wireless sensor networks, and geospatial big data management algorithms in the internet of manufacturing things. ISPRS Int J Geoinf2022;11(5):277. . 10.3390/ijgi11050277
[31]
LiM, XueF, WuY, YehAG. A room with a view: automatic assessment of window views for high-rise high-density areas using city information models and deep transfer learning. Landsc Urban Plan2022;226:104505. . 10.1016/j.landurbplan.2022.104505
[32]
ChanIY, LiuAM. Effects of neighborhood building density, height, greenspace, and cleanliness on indoor environment and health of building occupants. Build Environ2018;145:213‒22. . 10.1016/j.buildenv.2018.06.028
[33]
LeungM, WangC, ChanIY. A qualitative and quantitative investigation of effects of indoor built environment for people with dementia in care and attention homes. Build Environ2019;157:89‒100. . 10.1016/j.buildenv.2019.04.019
[34]
LuW, WuL, XuJ, LouJ. Construction E-inspection 2.0 in the COVID-19 pandemic era: a blockchain-based technical solution. J Manage Eng2022;38(4):04022032. . 10.1061/(asce)me.1943-5479.0001063
[35]
SolankiS, SolankiAD. Review of deployment of machine learning in blockchain methodology. Int Res J Adv Sci Hub2020;2(9):14‒20. . 10.47392/irjash.2020.141
[36]
ShindeR, NilakheO, PondkuleP, KarcheD, ShendageP. Enhanced road construction process with machine learning and blockchain technology. In: Proceedings of the 2020 International Conference on Industry 4.0 Technology (I4Tech); 2020 Feb 13‒15, Pune, India. New York City: IEEE; 2020. p. 207‒10. . 10.1109/i4tech48345.2020.9102669
[37]
WongP, ChiaF, KiuM, LouE. The potential of integrating blockchain technology into smart sustainable city development. In: Proceedings of the IOP Conference Series: Earth and Environmental Science; 2020 Sep 18‒20; Changsha, China. Bristol: IOP Publishing; 2020. p. 012020. . 10.1088/1755-1315/463/1/012020
[38]
XuJ, LuW, WuL, LouJ, LiX. Balancing privacy and occupational safety and health in construction: a blockchain-enabled P-OSH deployment framework. Saf Sci2022;154:105860. . 10.1016/j.ssci.2022.105860
[39]
RkK, KallapuB, DodmaneR, ThotaS, SahuAK. Enhancing cloud communication security: a blockchain-powered framework with attribute-aware encryption. Electronics (Basel)2023;12(18):3890. . 10.3390/electronics12183890
[40]
PodderD, DebS, BanikD, KarN, SahuAK. Robust medical and color image cryptosystem using array index and chaotic S-box. Cluster Comput2024;27(4):4321. . 10.1007/s10586-024-04584-3
[41]
JovanovicZ, HouZ, BiswasK, MuthukkumarasamyV. Robust integration of blockchain and explainable federated learning for automated credit scoring. Comput Netw2024;243:110303. . 10.1016/j.comnet.2024.110303
[42]
LiX, ZengJ, ChenC, ChiH, ShenGQ. Smart work package learning for decentralized fatigue monitoring through facial images. Comput Aided Civ Infrastruct Eng2023;38(6):799‒817. . 10.1111/mice.12891
[43]
KasyapH, TripathyS. Privacy-preserving and byzantine-robust federated learning framework using permissioned blockchain. Expert Syst Appl2024;238:122210. . 10.1016/j.eswa.2023.122210
[44]
WuL, LuW, ChenC. Strengths and weaknesses of client-server and peer-to-peer network models in construction projects. Int J Constr Manag2023;24(12):1349‒63. . 10.1080/15623599.2023.2185950
[45]
PeethambaranG, NaikodiC, SureshL. An ensemble learning approach for privacy-quality-efficiency trade-off in data analytics. In: Proceedings of the 2020 International Conference on Smart Electronics and Communication (ICOSEC); 2020 Sep 10‒12; Trichy, India. New York City: IEEE; 2020. p. 228‒35. . 10.1109/icosec49089.2020.9215250
[46]
CooperAF, LevyK, De SaC. Regulating accuracy-efficiency trade-offs in distributed machine learning systems. 2020. SSRN 3650497. . 10.1145/3465416.3483289
[47]
Al-MarridiAZ, MohamedA, ErbadA. Reinforcement learning approaches for efficient and secure blockchain-powered smart health systems. Comput Netw2021;197:108279. . 10.1016/j.comnet.2021.108279
[48]
JiangR, LiJ, BuW, ShenX. A blockchain-based trustworthy model evaluation framework for deep learning and its application in moving object segmentation. Sensors2023;23(14):6492. . 10.3390/s23146492
[49]
FengX, LiL, WangT, XuW, ZhangJ, WeiB, et al. CoBC: a blockchain-based collaborative inference system for the internet of things. IEEE Internet Things J2023;10(24):21389‒400. . 10.1109/jiot.2023.3290092
[50]
WangSJ, PeiK, YangJ. SmartInv: multimodal learning for smart contract invariant inference. In: Proceedings of the 2024 IEEESymposium on Security and Privacy (SP); 2024 May 19‒23; San Francisco, CA, USA. New York City: IEEE; 2024. . 10.1109/sp54263.2024.00126
[51]
AdelK, ElhakeemA, MarzoukM. Decentralizing construction ai applications using blockchain technology. Expert Syst Appl2022;194:116548. . 10.1016/j.eswa.2022.116548
[52]
AmershiS, BegelA, BirdC, DeLineR, GallH, KamarE, et al. Software engineering for machine learning: a case study. In: Proceedings of the 2019 IEEE/ACM 41st International Conference on Software Engineering: Software Engineering in Practice (ICSE-SEIP); 2019 May 25‒31; Montreal, QC, Canada. New York City: IEEE; 2019. p. 291‒300. . 10.1109/icse-seip.2019.00042
[53]
BertoliniM, MezzogoriD, NeroniM, ZammoriF. Machine learning for industrial applications: a comprehensive literature review. Expert Syst Appl2021;175:114820. . 10.1016/j.eswa.2021.114820
GerardC. Practical machine learning in JavaScript: TensorFlow.js for web developers. Berlin: Springer; 2021. p. 25‒43. . 10.1007/978-1-4842-6418-8_2
[56]
WuH, ZhongB, LiH, GuoJ, WangY. On-site construction quality inspection using blockchain and smart contracts. J Manage Eng2021;37(6):04021065. . 10.1061/(asce)me.1943-5479.0000967
[57]
OngaroD, OusterhoutJ. In search of an understandable consensus algorithm. In: Proceedings of the 2014 USENIX Annual Technical Conference (USENIX ATC 14); 2014 Jun 19‒20; Philadelphia, PA, USA. Berkeley: The USENIX Association; 2014. p. 305‒19.
HowardA, SandlerM, ChuG, ChenLC, ChenB, TanM, et al. Searching for mobilenetv3. In: Proceedings of the IEEE/CVF International Conference on Computer Vision; 2019 Oct 27‒Nov 2; Seoul, Republic of Korea. New York City: IEEE; 2019. p. 1314‒24. . 10.1109/iccv.2019.00140
[60]
DobbertinH. Cryptanalysis of MD5 compress. In: Proceedings of the EUROCRYPT’ 96 SessionRump; 1996 May 12‒16; Zaragoza, Spain. Bellevue: International Association for Cryptologic Research; 1996. p. 9671‒82.
[61]
AndersonJC, LehnardtJ, SlaterN. CouchDB: the definitive guide: time to relax. editors. Sebastopol: O’Reilly Media; 2010.
[62]
SavsunenkoO. How tensorflow’s tf. image. resize stole 60 days of my life [Internet]. Edwards: HackerNoon; 2018 Jan 23 [cited 2023 Dec 14]. Available from:
[63]
ZhengP, LiS, XiaL, WangL, NassehiA. A visual reasoning-based approach for mutual-cognitive human-robot collaboration. CIRP Ann2022;71(1):377‒80. . 10.1016/j.cirp.2022.04.016
[64]
GuoJ, LiuZ, TianS, HuangF, LiJ, LiX, et al. TFL-DT: a trust evaluation scheme for federated learning in digital twin for mobile networks. IEEE J Sel Areas Comm2023;41(11):3548‒60. . 10.1109/jsac.2023.3310094
[65]
MiaoJ, YangZ, FanL, YangY. FedSeg: class-heterogeneous federated learning for semantic segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition;2023 Jun 17‒24; Vancouver, BC, Canada. New York City: IEEE; 2023. p. 8042‒52. . 10.1109/cvpr52729.2023.00777
[66]
HuaH, LiY, WangT, DongN, LiW, CaoJ. Edge computing with artificial intelligence: a machine learning perspective. ACM Comput Surv2023;55(9):1‒35. . 10.1145/3555802
[67]
LiL, FanY, TseM, LinKY. A review of applications in federated learning. Comput Ind Eng2020;149:106854. . 10.1016/j.cie.2020.106854
[68]
KayikciS, KhoshgoftaarTM. Blockchain meets machine learning: a survey. J Big Data2024;11(1):9. . 10.1186/s40537-023-00852-y