Understanding the Complexity and Its Impact on Testing in ML-Enabled Systems

TL;DR

This paper investigates the complexity of large-scale ML-enabled systems like Rasa 3.0 and explores how this complexity influences testing practices, providing insights for software engineering in this domain.

Contribution

It adopts a system-level perspective to analyze the complexity of ML-enabled systems and its impact on testing, which is less explored compared to model-centric approaches.

Findings

Complexity in ML-enabled systems affects testing strategies.

System-level analysis reveals practical implications for software engineering.

Case study on Rasa 3.0 illustrates real-world challenges.

Abstract

Machine learning (ML) enabled systems are emerging with recent breakthroughs in ML. A model-centric view is widely taken by the literature to focus only on the analysis of ML models. However, only a small body of work takes a system view that looks at how ML components work with the system and how they affect software engineering for MLenabled systems. In this paper, we adopt this system view, and conduct a case study on Rasa 3.0, an industrial dialogue system that has been widely adopted by various companies around the world. Our goal is to characterize the complexity of such a largescale ML-enabled system and to understand the impact of the complexity on testing. Our study reveals practical implications for software engineering for ML-enabled systems.