Keep your Distance: Determining Sampling and Distance Thresholds in Machine Learning Monitoring

TL;DR

This paper introduces a practical approach for determining sampling sizes and distance thresholds in SafeML, a model-agnostic monitoring method for assessing the operational safety of ML components in variable environments.

Contribution

It provides a systematic method to set sampling and threshold parameters in SafeML, enhancing its reliability for real-world applications like autonomous driving.

Findings

Effective in traffic sign recognition task

Demonstrated on CARLA automotive simulator

Improves safety monitoring accuracy

Abstract

Machine Learning~(ML) has provided promising results in recent years across different applications and domains. However, in many cases, qualities such as reliability or even safety need to be ensured. To this end, one important aspect is to determine whether or not ML components are deployed in situations that are appropriate for their application scope. For components whose environments are open and variable, for instance those found in autonomous vehicles, it is therefore important to monitor their operational situation to determine its distance from the ML components' trained scope. If that distance is deemed too great, the application may choose to consider the ML component outcome unreliable and switch to alternatives, e.g. using human operator input instead. SafeML is a model-agnostic approach for performing such monitoring, using distance measures based on statistical testing of the training and operational datasets. Limitations in setting SafeML up properly include the lack of a systematic approach for determining, for a given application, how many operational samples are needed to yield reliable distance information as well as to determine an appropriate distance threshold. In this work, we address these limitations by providing a practical approach and demonstrate its use in a well known traffic sign recognition problem, and on an example using the CARLA open-source automotive simulator.