An investigation of challenges encountered when specifying training data and runtime monitors for safety critical ML applications

TL;DR

This study explores the key challenges practitioners face when specifying training data and runtime monitors for safety-critical ML systems, based on interviews with industry experts, and offers recommendations to address these issues.

Contribution

It identifies 17 underlying challenges and their interconnections in specifying training data and runtime monitoring for critical ML applications, providing targeted recommendations.

Findings

17 challenges identified across 6 groups

Interconnections between challenges discovered

Recommendations proposed to address root causes

Abstract

Context and motivation: The development and operation of critical software that contains machine learning (ML) models requires diligence and established processes. Especially the training data used during the development of ML models have major influences on the later behaviour of the system. Runtime monitors are used to provide guarantees for that behaviour. Question / problem: We see major uncertainty in how to specify training data and runtime monitoring for critical ML models and by this specifying the final functionality of the system. In this interview-based study we investigate the underlying challenges for these difficulties. Principal ideas/results: Based on ten interviews with practitioners who develop ML models for critical applications in the automotive and telecommunication sector, we identified 17 underlying challenges in 6 challenge groups that relate to the challenge of specifying training data and runtime monitoring. Contribution: The article provides a list of the identified underlying challenges related to the difficulties practitioners experience when specifying training data and runtime monitoring for ML models. Furthermore, interconnection between the challenges were found and based on these connections recommendation proposed to overcome the root causes for the challenges.