Machine Learning Operations aplicado al proceso de desarrollo y aprovisionamiento de modelos

dc.contributor.advisorCamargo Mendoza, Jorge Eliécerspa
dc.contributor.advisorFlórez Fernández, Héctor Arturospa
dc.contributor.authorMendez Aguirre, Oscar Alexanderspa
dc.contributor.researchgroupUnSecureLabspa
dc.date.accessioned2024-05-06T20:27:00Z
dc.date.available2024-05-06T20:27:00Z
dc.date.issued2024
dc.descriptionilustraciones, diagramasspa
dc.description.abstractEn la actual era de la ingeniería de software, donde el Machine Learning (ML) desempeña un papel crucial en la innovación tecnológica, la aplicación efectiva de prácticas de desarrollo y operación es esencial. El enfoque de DevSecOps (Development Security Operations) se ha popularizado por su capacidad para integrar la seguridad y la calidad en todas las etapas del ciclo de vida del desarrollo seguro de software. Sin embargo, en el contexto específico del Machine Learning, surge la necesidad de un enfoque especializado que considere las particula- ridades de los modelos y algoritmos utilizados. El Machine Learning Operations (MLOps), a pesar de su relativa novedad, busca establecer un marco para caracterizar el ciclo de vida del desarrollo de ML, desacoplarlo del desarrollo de software y garantizar atributos de calidad como escalabilidad, mantenibilidad y seguridad. También se enfrenta al desafío de gestionar datos de entrenamiento, la seguridad en el proceso de análisis y desarrollo de modelos, y la necesidad de una cultura orientada a la calidad. Este trabajo se centra en investigar cómo la implementación de MLOps puede impactar positivamente en la gestión del ciclo de vida del desarrollo de ML, con el objetivo de contribuir al conocimiento en este campo emergente y promover la adopción de las mejores prácticas en soluciones basadas en ML. (Texto tomado de la fuente).spa
dc.description.abstractIn the current era of software engineering, where Machine Learning (ML) plays a pivotal role in technological innovation, the effective implementation of development and opera- tions practices is essential. The DevSecOps (Development Security Operations) approach has gained popularity due to its ability to integrate security and quality at every stage of the software development lifecycle. However, in the specific context of Machine Learning, there arises a need for a specialized approach that takes into account the peculiarities of the models and algorithms used. Machine Learning Operations (MLOps), despite its relative immaturity, aims to establish a framework for characterizing the ML development lifecycle, decoupling it from software development, and ensuring quality attributes such as scalability, maintainability, and security. It also grapples with challenges related to managing training data, security throughout the model analysis, development and deployment process, and the need for a quality-oriented culture. This thesis focuses on investigating how the implementa- tion of MLOps can positively impact the management of the ML development lifecycle, with the goal of contributing to knowledge in this emerging field and promoting the adoption of best practices in ML-based solutions.eng
dc.description.degreelevelMaestríaspa
dc.description.degreenameMagíster en Ingeniería - Ingeniería de Sistemas y Computaciónspa
dc.description.researchareaAplicaciones del machine learning operationsspa
dc.format.extentxiv, 103 páginasspa
dc.format.mimetypeapplication/pdfspa
dc.identifier.instnameUniversidad Nacional de Colombiaspa
dc.identifier.reponameRepositorio Institucional Universidad Nacional de Colombiaspa
dc.identifier.repourlhttps://repositorio.unal.edu.co/spa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/86035
dc.language.isospaspa
dc.publisherUniversidad Nacional de Colombiaspa
dc.publisher.branchUniversidad Nacional de Colombia - Sede Bogotáspa
dc.publisher.facultyFacultad de Ingenieríaspa
dc.publisher.placeBogotá, Colombiaspa
dc.publisher.programBogotá - Ingeniería - Maestría en Ingeniería - Ingeniería de Sistemas y Computaciónspa
dc.relation.referencesAmershi, S., Begel, A., Bird, C., DeLine, R., Gall, H., Kamar, E., Maddox, W., Maarek, Y., McDirmid, S., New, M., et al. (2022). Software engineering for machine learning: An experience report. IEEE Software, 39(5):68–75.spa
dc.relation.referencesBanerjee, A., Chen, C.-C., Hung, C.-C., Huang, X., Wang, Y., and Chevesaran, R. (2020). Challenges and experiences with mlops for performance diagnostics in hybrid-cloud enter- prise software deployments.spa
dc.relation.referencesBaylor, D., Haas, K., Katsiapis, K., Leong, S., Liu, R., Menwald, C., Miao, H., Polyzotis, N., Trott, M., and Zinkevich, M. (2019). Continuous training for production tMLu in the tTensorFlowu extended (tttttTFXuuuuu) platform. In 2019 USENIX Conference on Operational Machine Learning (OpML 19), pages 51–53.spa
dc.relation.referencesBenjumea, J., Ropero, J., Rivera, J., Mavrogiannopoulos, N., and Kamhoua, C. (2022). Catalog of requirements for artificial intelligence products. Future Internet, 14(5):118.spa
dc.relation.referencesBodor, A., Hnida, M., and Najima, D. (2023). MLOps: Overview of Current State and Future Directions, volume 629 LNNS.spa
dc.relation.referencesChatterjee, A., Ahmed, B., Hallin, E., and Engman, A. (2022a). Quality assurance in mlops setting: An industrial perspective. volume 3362.spa
dc.relation.referencesChatterjee, A., Ahmed, B. S., Hallin, E., and Engman, A. (2022b). Quality assurance in mlops setting: An industrial perspective. arXiv preprint arXiv:2211.12706.spa
dc.relation.referencesDíaz, J., Pérez, J. E., Lopez-Peña, M. A., Mena, G. A., and Yagüe, A. (2019). Self-service cybersecurity monitoring as enabler for devsecops. Ieee Access, 7:100283–100295.spa
dc.relation.referencesDrivenData (2023). Cookiecutter data science. https://drivendata.github.io/cookiecutter- data-science/. Último acceso: [Fecha de último acceso].spa
dc.relation.referencesEck, B., Kabakci-Zorlu, D., Chen, Y., Savard, F., and Bao, X. (2022). A monitoring frame- work for deployed machine learning models with supply chain examples. arXiv preprint arXiv:2211.06239.spa
dc.relation.referencesFujii, T. Y., Hayashi, V. T., Arakaki, R., Ruggiero, W. V., Bulla Jr, R., Hayashi, F. H., and Khalil, K. A. (2021). A digital twin architecture model applied with mlops techniques to improve short-term energy consumption prediction. Machines, 10(1):23.spa
dc.relation.referencesGartner (2016). Gartner Says Business Intelligence and Analytics Leaders Must Focus on Mindsets and Culture to Kick Start Advanced Analytics. Technical report, Gartner.spa
dc.relation.referencesGarzas, J. and Piattini, M. (2022). Failure factors in machine learning projects. In Advances in Intelligent Systems and Computing, volume 1197, pages 21–31. Springer.spa
dc.relation.referencesGhanta, S., Subramanian, S., Khermosh, L., Sundararaman, S., Shah, H., Goldberg, Y., Roselli, D., and Talagala, N. (2019). Ml health monitor: Taking the pulse of machine learning algorithms in production. volume 11139.spa
dc.relation.referencesGoogle Cloud (2022). Mlops: Continuous delivery and automation pipelines in machine learning.spa
dc.relation.referencesGärtler, M., Khaydarov, V., Klöpper, B., and Urbas, L. (2021). The machine learning life cycle in chemical operations – status and open challenges. Chemie-Ingenieur-Technik, 93:2063–2080.spa
dc.relation.referencesHernandez, J., Daza, K., and Florez, H. (2022). Spiking neural network approach based on caenorhabditis elegans worm for classification. IAENG International Journal of Computer Science, 49(4).spa
dc.relation.referencesHewage, N. and Meedeniya, D. (2022). Machine learning operations: A survey on mlops tool support. arXiv.spa
dc.relation.referencesHumble, J. and Kim, G. (2018). Accelerate: The science of lean software and devops: Building and scaling high performing technology organizations. IT Revolution.spa
dc.relation.referencesJohn, M., Olsson, H., and Bosch, J. (2021). Towards mlops: A framework and maturity model. pages 334–341.spa
dc.relation.referencesKitchenham, B. and Brereton, P. (2013). A systematic review of systematic review process research in software engineering. Information and software technology, 55(12):2049–2075.spa
dc.relation.referencesKreuzberger, D., Kühl, N., and Hirschl, S. (2023). Machine learning operations (mlops): Overview, definition, and architecture. IEEE Access.spa
dc.relation.referencesLim, J., Lee, H., Won, Y., and Yeon, H. (2019). Mlop lifecycle scheme for vision-based inspection process in manufacturing. pages 9–11.spa
dc.relation.referencesLima, A., Monteiro, L., and Furtado, A. (2022). Mlops: Practices, maturity models, roles, tools, and challenges - a systematic literature review. volume 1, pages 308–320.spa
dc.relation.referencesLiu, L. T., Wang, S., Britton, T., and Abebe, R. (2023). Reimagining the machine learning life cycle to improve educational outcomes of students. Proceedings of the National Academy of Sciences, 120(9):e2204781120.spa
dc.relation.referencesMakinen, S., Skogstrom, H., Laaksonen, E., and Mikkonen, T. (2021). Who needs mlops: What data scientists seek to accomplish and how can mlops help? pages 109–112.spa
dc.relation.referencesMalhotra, Y. (2022). How you can implement well-architected ‘zero trust’hybrid-cloud com- puting beyond ‘lift and shift’: cloud-enabled digital innovation at scale with infrastructure as code (iac), devsecops and mlops. In 2022 New York State Cyber Security Conference: Invited Presentations, Albany, New York: https://its. ny. gov/2022-nyscsc.spa
dc.relation.referencesMartel, Y., Roßmann, A., Sultanow, E., Weiß, O., Wissel, M., Pelzel, F., and Seßler, M. (2021). Software architecture best practices for enterprise artificial intelligence. INFOR- MATIK 2020.spa
dc.relation.referencesMboweni, T., Masombuka, T., and Dongmo, C. (2022). A systematic review of machine learning devops. In 2022 International Conference on Electrical, Computer and Energy Technologies (ICECET), pages 1–6. IEEE.spa
dc.relation.referencesMejía, J. A. G. and González, F. (2022). Encuesta nacional “machine learning operations y sus desafíos de implementación en colombia”. Revista Sistemas, (165):20–26.spa
dc.relation.referencesMirza, B., Li, X., Lauwers, K., Reddy, B., Muller, A., Wozniak, C., and Djali, S. (2023). A clinical site workload prediction model with machine learning lifecycle. Healthcare Analytics, 3:100159.spa
dc.relation.referencesParihar, A. S., Gupta, U., Srivastava, U., Yadav, V., and Trivedi, V. K. (2023). Automa- ted machine learning deployment using open-source ci/cd tool. In Proceedings of Data Analytics and Management: ICDAM 2022, pages 209–222. Springer.spa
dc.relation.referencesPaul, S. K., Riaz, S., and Das, S. (2022). A conceptual architecture for ai in supply chain risk management. In TENCON 2022-2022 IEEE Region 10 Conference (TENCON), pages 1–5. IEEE.spa
dc.relation.referencesRecupito, G., Pecorelli, F., Catolino, G., Moreschini, S., Nucci, D., Palomba, F., and Tam- burri, D. (2022). A multivocal literature review of mlops tools and features. pages 84–91.spa
dc.relation.referencesReddy, M., Dattaprakash, B., Kammath, S., Kn, S., Manokaran, S., and Be, R. (2022). Application of mlops in prediction of lifestyle diseases. ECS Transactions, 107(1):1191.spa
dc.relation.referencesRobertson, J. and Robertson, S. (2000). Volere. Requirements Specification Templates.spa
dc.relation.referencesSaeed, W. and Omlin, C. (2023). Explainable ai (xai): A systematic meta-survey of current challenges and future opportunities. Knowledge-Based Systems, page 110273.spa
dc.relation.referencesSchulz, C., Gao, J., and Sun, T. (2020). Mlops: A literature review. In 2020 IEEE 1st International Workshop on Machine Learning in Business Process Management (BPML). IEEE.spa
dc.relation.referencesSculley, D., Holt, G., Golovin, D., Davydov, E., Phillips, T., Ebner, D., Chaudhary, V., Young, M., Crespo, J.-F., and Dennison, D. (2015). Hidden technical debt in machine learning systems. Advances in neural information processing systems, 28.spa
dc.relation.referencesTreveil, M., Omont, N., Stenac, C., Lefevre, K., Phan, D., Zentici, J., Lavoillotte, A., Miya- zaki, M., and Heidmann, L. (2020). Introducing MLOps. O’Reilly Media.spa
dc.relation.referencesYasar, H. (2020). Leveraging devops and devsecops to accelerate ai development and deploy- ment. CARNEGIEMELLON UNIV PITTSBURGH PA PITTSBURGH United States.spa
dc.relation.referencesZhang, X. and Jaskolka, J. (2022). Conceptualizing the secure machine learning operations (secmlops) paradigm. In 2022 IEEE 22nd International Conference on Software Quality, Reliability and Security (QRS), pages 127–138. IEEE.spa
dc.relation.referencesZhou, Y., Yu, Y., and Ding, B. (2020). Towards mlops: A case study of ml pipeline platform. pages 494–500.spa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.licenseAtribución-NoComercial 4.0 Internacionalspa
dc.rights.urihttp://creativecommons.org/licenses/by-nc/4.0/spa
dc.subject.ddc620 - Ingeniería y operaciones afines::629 - Otras ramas de la ingenieríaspa
dc.subject.ddc000 - Ciencias de la computación, información y obras generales::005 - Programación, programas, datos de computaciónspa
dc.subject.proposalMachine Learningspa
dc.subject.proposalMLOpsspa
dc.subject.proposalDevSecOpsspa
dc.subject.proposalGestión de datosspa
dc.subject.proposalInnovación tecnológicaspa
dc.subject.proposalDesarrollo de softwarespa
dc.subject.proposalMachine Learningeng
dc.subject.proposalMLOpseng
dc.subject.proposalDevSecOpseng
dc.subject.proposalSecurityeng
dc.subject.proposalData managementeng
dc.subject.proposalTechnological innovationeng
dc.subject.proposalSoftware developmenteng
dc.subject.wikidataAprendizaje automáticospa
dc.subject.wikidatamachine learningeng
dc.subject.wikidataIntegridad de datosspa
dc.subject.wikidatadata integrityeng
dc.subject.wikidataGestión de datosspa
dc.subject.wikidatadata managementeng
dc.titleMachine Learning Operations aplicado al proceso de desarrollo y aprovisionamiento de modelosspa
dc.title.translatedMachine Learning Operations applied to the process of model development and provisioningeng
dc.typeTrabajo de grado - Maestríaspa
dc.type.coarhttp://purl.org/coar/resource_type/c_bdccspa
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aaspa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/masterThesisspa
dc.type.redcolhttp://purl.org/redcol/resource_type/TMspa
dc.type.versioninfo:eu-repo/semantics/acceptedVersionspa
dcterms.audience.professionaldevelopmentPúblico generalspa
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2spa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
80121552.2024.pdf
Tamaño:
6.68 MB
Formato:
Adobe Portable Document Format
Descripción:
Tesis de Maestría en Ingeniería - Ingeniería de Sistemas y Computación

Bloque de licencias

Mostrando 1 - 1 de 1
No hay miniatura disponible
Nombre:
license.txt
Tamaño:
5.74 KB
Formato:
Item-specific license agreed upon to submission
Descripción: