Automated Explanation Selection for Scientific Discovery

TL;DR

This paper introduces a cycle of scientific discovery combining machine learning and automated reasoning to generate and select explanations, enhancing explainability in AI systems.

Contribution

It proposes a new taxonomy of explanation selection problems, integrating insights from sociology and cognitive science, and extending existing criteria.

Findings

Developed a taxonomy of explanation selection problems

Extended existing explanation criteria with new properties

Bridged insights from sociology and cognitive science

Abstract

Automated reasoning is a key technology in the young but rapidly growing field of Explainable Artificial Intelligence (XAI). Explanability helps build trust in artificial intelligence systems beyond their mere predictive accuracy and robustness. In this paper, we propose a cycle of scientific discovery that combines machine learning with automated reasoning for the generation and the selection of explanations. We present a taxonomy of explanation selection problems that draws on insights from sociology and cognitive science. These selection criteria subsume existing notions and extend them with new properties.