The AIDA project partners presented a paper at the 1st International Workshop on Software Architecture and Machine Learning (SAML) 2021, which is co-located with the 15th European Conference on Software Architecture (ECSA) 2021 and took place online on September 14, 2021.

Today’s world is witnessing the rise of systems that rely on machine learning. These systems typically operate in environments that are prone to unexpected changes, as is the case of self-driving cars and enterprise systems. Due to the unexpected changes, machine-learned software in these systems can malfunction. Thus, it is paramount that these systems are capable of detecting problems with their machined-learned components and adapt themselves to maintain desired qualities. For instance, a fraud detection system that cannot adapt its machine-learned model to efficiently cope with emerging fraud patterns or changes in the volume of transactions is subject to losses of millions of dollars. In this work, the authors take a first step towards the development of a framework aimed at self-adapting systems that rely on machine-learned components.

The article entitled “Self-Adaptation for Machine Learning Based Systems” and concludes with a set of research questions to guide future work. 

The author of this paper from AIDA is David Garlan from Carnegie Mellon University and Maria Casimiro, a CMU Portugal PhD student.