Software Engineering Patterns for Machine Learning Applications (SEP4MLA) - Part 4 - ML Gateway Routing Architecture. Washizaki, H., Khomh, F., & Gu�h�neuc, Y. In Manns, M. L. & Guerra, E., editors, Proceedings of the 29<sup>th</sup> Conference on Pattern Languages of Programs (PLoP), pages 1&ndash;10, October, 2022. ACM Press. 10 pages.
Software Engineering Patterns for Machine Learning Applications (SEP4MLA) - Part 4 - ML Gateway Routing Architecture [pdf]Paper  abstract   bibtex   
Machine learning (ML) researchers study the best practices to develop and support ML-based applications to ensure quality and determine the constraints applied to their application pipelines. Such practices are often formalized as software patterns. We discovered software-engineering design patterns for machine-learning applications by thoroughly searching the available literature on the subject. Among the ML patterns found, we describe in this paper one ML topology pattern, "ML Gateway Routing Architecture", in the standard pattern format so that practitioners can (re)use it in their contexts and benefits. The pattern addresses the problem of tight coupling among ML-implemented and non-ML business logic as well as the front-end client by installing a gateway that routes requests.

Downloads: 0