Упровадження практик MLOps для ефективного розгортання моделей машинного навчання

Ганчук, Данило Олегович

URI: http://elibrary.kdpu.edu.ua/xmlui/handle/123456789/10242

Date: 2024-06-12

Abstract:

У квалiфiкацiйнiй роботi розглянуто проблему впровадження практик MLOps для ефективного розгортання моделей машинного навчання. Виконано мета-синтез систематичних оглядiв для узагальнення знань щодо практик MLOps. Проаналiзовано зв'язки мiж принципами, процесами та практиками MLOps. Виявлено найбiльш ефективнi практики MLOps для розгортання моделей машинного навчання, такi як безперервна iнтеграцiя та доставка, версiонування моделей та даних, автоматизацiя конвеєрiв, монiторинг продуктивностi, управлiння життєвим циклом. Запропоновано схему зв'язкiв мiж принципами, процесами та практиками MLOps. Результати роботи можуть бути використанi органiзацiями для впровадження або вдосконалення процесiв MLOps з метою пiдвищення ефективностi та надiйностi розгортання моделей машинного навчання у виробничих середовищах.

Description:

1. “If security is required”: engineering and security practices for machine learning-based IoT devices / N. K. Gopalakrishna [та ін.] // Proceedings of the 4th International Workshop on Software Engineering Research and Practice for the IoT. — Pittsburgh, Pennsylvania : Association for Computing Machinery, 2023. — С. 1—8. — (SERP4IoT ’22). — DOI: 10.1145/3528227.3528565. 2. A Joint Study of the Challenges, Opportunities, and Roadmap of MLOps and AIOps: A Systematic Survey / J. Diaz-de-Arcaya [та ін.] // ACM Comput. Surv. — New York, NY, USA, 2023. — Жовт. — Т. 56, № 4. — DOI: 10.1145/3625289. 3. A Multivocal Literature Review of MLOps Tools and Features / G. Recupito [та ін.] // 2022 48th Euromicro Conference on Software Engineering and Advanced Applications (SEAA). — 2022. — С. 84—91. — DOI: 10.1109/SEAA56994.2022.00021. 4. A Software Ecosystem for Deploying Deep Learning in Gravitational Wave Physics / A. Gunny [та ін.] // Proceedings of the 12th Workshop on AI and Scientific Computing at Scale Using Flexible Computing Infrastructures. — Minneapolis, MN, USA : Association for Computing Machinery, 2022. — С. 9—17. — (FlexScience ’22). — DOI: 10.1145/3526058.3535454. 5. Automating Tiny ML Intelligent Sensors DevOPS Using Microsoft Azure / C. Vuppalapati [та ін.] // 2020 IEEE International Conference on Big Data (Big Data). — 2020. — С. 2375—2384. — DOI: 10.1109/BigData50022.2020.9377755. 6. Bachinger F., Zenisek J., Affenzeller M. Automated Machine Learning for Industrial Applications – Challenges and Opportunities // Procedia Computer Science. — 2024. — Т. 232. — С. 1701—1710. — DOI: https://doi.org/10.1016/j.procs.2024.01.168. 7. Bodor A., Hnida M., Daoudi N. Machine Learning Models Monitoring in MLOps Context: Metrics and Tools // International Journal of Interactive 81Mobile Technologies (iJIM). — 2023. — Груд. — Т. 17, № 23. — pp. 125—139. — DOI: 10.3991/ijim.v17i23.43479. 8. Building Network Domain Knowledge Graph from Heterogeneous YANG Models / D. Yongqiang [та ін.] // Journal of Computer Research and Development. — 2020. — Т. 57, № 4. — С. 699—708. — DOI: 10.7544/issn1000-1239.2020.20190882. 9. Calefato F., Lanubile F., Quaranta L. A Preliminary Investigation of MLOps Practices in GitHub // Proceedings of the 16th ACM / IEEE International Symposium on Empirical Software Engineering and Measurement. — Helsinki, Finland : Association for Computing Machinery, 2022. — С. 283—288. — (ESEM ’22). — DOI: 10.1145/3544902.3546636. 10. Characterizing Machine Learning Processes: A Maturity Framework / R. Akkiraju [та ін.] // Business Process Management. Т. 12168 / за ред. D. Fahland [та ін.]. — Cham : Springer International Publishing, 2020. — С. 17—31. — (Lecture Notes in Computer Science). — ISBN 978-3-030-58666-9. — DOI: 10.1007/978-3-030-58666-9_2. 11. Chen H., Babar M. A. Security for Machine Learning-based Software Systems: A Survey of Threats, Practices, and Challenges // ACM Comput. Surv. — New York, NY, USA, 2024. — Лют. — Т. 56, № 6. — DOI: 10.1145/3638531. 12. Chrastina J. Meta-synthesis of qualitative studies: background, methodology and applications // NORDSCI Conference proceedings. Т. 1. — Saima Consult Ltd, 2018. — (NORDSCI Conference). — DOI: 10.32008/nordsci2018/b1/v1/13. 13. Cohen R. Digital Strategy, Machine Learning, and Industry Survey of MLOps // Digital Strategies and Organizational Transformation. — 2023. — Гл. 8. С. 137—150. — DOI: 10.1142/9789811271984_0008. — URL: https://tinyurl.com/33z6zpd3. 14. Czakon J., Kluge K. ML Experiment Tracking: What It Is, Why It Matters, and How to Implement It. — 05.2024. — URL: https://neptune.ai/blog/ml-experiment-tracking. 8215. Ease.ML: A Lifecycle Management System for MLDev and MLOps / L. A. Melgar [та ін.] // 11th Conference on Innovative Data Systems Research, CIDR 2021, Virtual Event, January 11-15, 2021, Online Proceedings. — 2021. — URL: https://www.cidrdb.org/cidr2021/papers/cidr2021_paper26.pdf. 16. Godwin R. C., Melvin R. L. Toward efficient data science: A comprehensive MLOps template for collaborative code development and automation // SoftwareX. — 2024. — Т. 26. — DOI: 10.1016/j.softx.2024.101723. 17. Haller K. Managing AI in the enterprise: Succeeding with AI projects and MLOps to build sustainable AI organizations. — 2022. — С. 1—214. — DOI: 10.1007/978-1-4842-7824-6. 18. Kolltveit A. B., Li J. Operationalizing machine learning models: a systematic literature review // Proceedings of the 1st Workshop on Software Engineering for Responsible AI. — Pittsburgh, Pennsylvania : Association for Computing Machinery, 2023. — С. 1—8. — (SE4RAI ’22). — DOI: 10.1145/3526073.3527584. 19. Kreuzberger D., Kühl N., Hirschl S. Machine Learning Operations (MLOps): Overview, Definition, and Architecture // IEEE Access. — 2023. — Т. 11. — С. 31866—31879. — DOI: 10.1109/ACCESS.2023.3262138. 20. Lima A., Monteiro L., Furtado A. P. MLOps: Practices, Maturity Models, Roles, Tools, and Challenges – A Systematic Literature Review // Proceedings of the 24th International Conference on Enterprise Information Systems - Volume 1: ICEIS. — INSTICC. SciTePress, 2022. — С. 308—320. — ISBN 978-989-758-569-2. — DOI: 10.5220/0010997300003179. 21. Lwakatare L. E., Crnkovic I., Bosch J. DevOps for AI – Challenges in Development of AI-enabled Applications // 2020 International Conference on Software, Telecommunications and Computer Networks (SoftCOM). — 2020. — С. 1—6. — DOI: 10.23919/SoftCOM50211.2020.9238323. 22. MLOps - Definitions, Tools and Challenges / G. Symeonidis [та ін.] // 2022 IEEE 12th Annual Computing and Communication Workshop and Conference (CCWC). — 2022. — С. 0453—0460. — DOI: 10.1109/CCWC54503.2022.9720902. 8323. MLOps in Data Science Projects: A Review / C. Haertel [та ін.] // 2023 IEEE International Conference on Big Data (BigData). — 2023. — С. 2396—2404. — DOI: 10.1109/BigData59044.2023.10386139. 24. MLOps: A Taxonomy and a Methodology / M. Testi [та ін.] // IEEE Access. — 2022. — Т. 10. — С. 63606—63618. — DOI: 10.1109/ACCESS.2022.3181730. 25. Neptune Labs. MLOps Landscape in 2024: Top Tools and Platforms. — 2024. — URL: https://neptune.ai/blog/mlops-tools-platforms- landscape. 26. Peltonen E., Dias S. LinkEdge: Open-sourced MLOps Integration with IoT Edge // Proceedings of the 3rd Eclipse Security, AI, Architecture and Modelling Conference on Cloud to Edge Continuum. — Ludwigsburg, Germany : Association for Computing Machinery, 2023. — С. 67—76. — (ESAAM ’23). — DOI: 10.1145/3624486.3624496. 27. SensiX++: Bringing MLOps and Multi-tenant Model Serving to Sensory Edge Devices / C. Min [та ін.] // ACM Trans. Embed. Comput. Syst. — New York, NY, USA, 2023. — Листоп. — Т. 22, № 6. — DOI: 10.1145/3617507. — URL: https://doi.org/10.1145/3617507. 28. Singh P. Systematic review of data-centric approaches in artificial intelligence and machine learning // Data Science and Management. — 2023. — Т. 6, № 3. — С. 144—157. — DOI: https://doi.org/10.1016/j.dsm.2023.06.001. 29. Sipe T. A., Curlette W. L. A meta-synthesis of factors related to educational achievement: a methodological approach to summarizing and synthesizing meta-analyses // International Journal of Educational Research. — 1996. — Т. 25, № 7. — С. 583—698. — DOI: 10.1016/S0883-0355(96)80001-2. 30. SliceOps: Explainable MLOps for Streamlined Automation-Native 6G Networks / F. Rezazadeh [та ін.] // IEEE Wireless Communications. — 2024. — С. 1—7. — DOI: 10.1109/MWC.007.2300144. 8431. Software Engineering for Machine Learning: A Case Study / S. Amershi [та ін.] // 2019 IEEE/ACM 41st International Conference on Software Engineering: Software Engineering in Practice (ICSE-SEIP). — 2019. — С. 291—300. — DOI: 10.1109/ICSE-SEIP.2019.00042. 32. Sothilingam R., Pant V., Yu E. S. K. Using i* to Analyze Collaboration Challenges in MLOps Project Teams // Proceedings of the 15th International iStar Workshop (iStar 2022) co-located with 41th International Conference on Conceptual Modeling (ER 2022), Virtual Event, Hyderabad, India, October 17, 2022. Т. 3231 / за ред. A. Maté, T. Li, E. J. T. Gonçalves. — CEUR-WS.org, 2022. — С. 1—6. — (CEUR Workshop Proceedings). — URL: https://ceur- ws.org/Vol- 3231/ iStar22%5C_paper%5C_1.pdf. 33. Steidl M., Felderer M., Ramler R. The pipeline for the continuous development of artificial intelligence models—Current state of research and practice // Journal of Systems and Software. — 2023. — Т. 199. — С. 111615. — DOI: 10.1016/j.jss.2023.111615. 34. Structure Learning and Hyperparameter Optimization Using an Automated Machine Learning (AutoML) Pipeline / K. Filippou [та ін.] // Information. — 2023. — Т. 14, № 4. — С. 232. — DOI: 10 . 3390 / info14040232. 35. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews / M. J. Page [та ін.] // BMJ. — 2021. — Т. 372. — n71. — DOI: 10.1136/bmj.n71. 36. Unlabeled learning algorithms and operations: overview and future trends in defense sector / E. e Oliveira [та ін.] // Artificial Intelligence Review. — 2024. — Т. 57, № 3. — DOI: 10.1007/s10462-023-10692-0. 37. Who needs to know what, when?: Broadening the Explainable AI (XAI) Design Space by Looking at Explanations Across the AI Lifecycle / S. Dhanorkar [та ін.] // Proceedings of the 2021 ACM Designing Interactive Systems Conference. — Virtual Event, USA : Association for Computing Machinery, 2021. — С. 1591—1602. — (DIS ’21). — DOI: 10.1145/3461778.3462131.

Show full item record