Model Agnostic Local Explanations of Reject

TL;DR

This paper introduces a model-agnostic approach to locally explain reject options in machine learning systems, using interpretable models and counterfactuals to clarify why samples are rejected, which is crucial for safety-critical applications.

Contribution

It presents a novel method for explaining reject decisions in machine learning, addressing the open problem of interpretability for reject options.

Findings

Enables local explanations for reject decisions

Uses interpretable models and counterfactuals for explanations

Applicable to various reject options in ML systems

Abstract

The application of machine learning based decision making systems in safety critical areas requires reliable high certainty predictions. Reject options are a common way of ensuring a sufficiently high certainty of predictions made by the system. While being able to reject uncertain samples is important, it is also of importance to be able to explain why a particular sample was rejected. However, explaining general reject options is still an open problem. We propose a model agnostic method for locally explaining arbitrary reject options by means of interpretable models and counterfactual explanations.