technical debt

Hidden technical debt in machine learning systems (paper summary)

Posted on November 30, 2021

Machine learning systems are wonderful. Many shapes and forms of machine learning algorithms are currently in use. Different models such as clustering like k-means, prediction methods like trees, or more advanced deep learning methods suffer from technical debt. In traditional software engineering, technical debt can be found in specific shapes. In addition to the “traditional” software engineering problems, machine learning systems also face new challenges. The following paragraphs present the different technical debt found in machine learning systems. 1. EncapsulationRead More