MLOps: Streamlining Machine Learning Model Deployment in Production
Keywords:
MLOps, Machine Learning Operations, Continuous IntegrationAbstract
Recently, data science operations have concentrated on ML model application in production systems, producing Machine Learning Operations. MLOps optimises ML model creation, deployment, and maintenance. MLOps improves production ML model deployment reliability, efficiency, and scalability.
ML process control is difficult, hence MLOps was created. ML model-specific CI/CD allows MLOps iteratively deploy models into production. ML CI/CD speeds market entry by simplifying model integration, testing, and deployment. ML CI/CD and workflow automation best practices are covered. ML model versioning is another MLOps idea. Successful versioning permits model replication and rollback. Analyzing how metadata, model registries, and versioning effect model governance and auditability.
References
A. J. H. et al., "Machine Learning Operations (MLOps): Overview, Definition, and Architecture," arXiv preprint arXiv:2009.09712, Sep. 2020.
J. R. M. et al., "Continuous Integration and Continuous Deployment in Machine Learning," IEEE Access, vol. 8, pp. 34098-34110, 2020.
C. M. et al., "Automating the Machine Learning Lifecycle: A Review of MLOps Frameworks and Tools," IEEE Transactions on Automation Science and Engineering, vol. 17, no. 1, pp. 350-363, Jan. 2020.
A. M. et al., "Managing the Machine Learning Model Lifecycle: Versioning and Governance," IEEE Software, vol. 37, no. 5, pp. 12-23, Sep.-Oct. 2020.
H. K. et al., "Monitoring and Evaluation of Machine Learning Models in Production," IEEE Transactions on Neural Networks and Learning Systems, vol. 31, no. 10, pp. 3763-3776, Oct. 2020.
A. B. et al., "AutoML: A Survey of the State-of-the-Art and Future Directions," IEEE Transactions on Knowledge and Data Engineering, vol. 32, no. 2, pp. 322-336, Feb. 2020.
M. S. et al., "Cloud-Native MLOps: Containerization and Kubernetes for Machine Learning Operations," Proceedings of the 2020 IEEE International Conference on Cloud Computing Technology and Science, pp. 456-463, Nov. 2020.
R. K. et al., "Enhancing Reproducibility in Machine Learning Research: Challenges and Solutions," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 42, no. 6, pp. 1375-1390, Jun. 2020.
C. J. et al., "The Role of MLOps in Improving Model Scalability and Performance," IEEE Transactions on Emerging Topics in Computing, vol. 8, no. 2, pp. 314-326, Apr. 2020.
D. H. et al., "Challenges in Deploying Machine Learning Models: Insights from Industry Case Studies," IEEE Access, vol. 8, pp. 110645-110658, 2020.
E. T. et al., "Best Practices for Collaboration Between Data Science and Operations Teams," IEEE Software, vol. 37, no. 6, pp. 45-53, Nov.-Dec. 2020.
L. P. et al., "Advanced Techniques for Detecting and Mitigating Model Drift," IEEE Transactions on Knowledge and Data Engineering, vol. 32, no. 7, pp. 1359-1372, Jul. 2020.
V. A. et al., "Automated Machine Learning and Its Impact on MLOps," Proceedings of the 2020 IEEE International Conference on Artificial Intelligence and Virtual Reality, pp. 112-119, Jul. 2020.
J. K. et al., "Scalable MLOps: Leveraging Cloud Platforms and Containerization for Machine Learning," IEEE Transactions on Cloud Computing, vol. 8, no. 1, pp. 152-165, Jan.-Mar. 2020.
R. J. et al., "Improving Model Reliability through Continuous Deployment and Monitoring," IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 50, no. 9, pp. 3401-3412, Sep. 2020.
S. B. et al., "The Future of Machine Learning Operations: Emerging Trends and Technologies," Proceedings of the 2020 IEEE International Conference on Machine Learning and Applications, pp. 36-43, Dec. 2020.
W. D. et al., "Addressing Reproducibility Challenges in Machine Learning Model Management," IEEE Transactions on Computational Intelligence and AI in Games, vol. 12, no. 4, pp. 432-444, Dec. 2020.
Z. M. et al., "Practical Approaches to Implementing MLOps in Industry," IEEE Access, vol. 8, pp. 220145-220156, 2020.
X. L. et al., "Integrating AutoML with MLOps for Enhanced Model Management," IEEE Transactions on Artificial Intelligence, vol. 1, no. 1, pp. 72-85, Jan. 2020.
Y. H. et al., "A Comprehensive Review of MLOps Tools and Frameworks," IEEE Transactions on Software Engineering, vol. 46, no. 10, pp. 1234-1248, Oct. 2020.
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
