Explainable ASP

摘要

Despite its proven relevance, ASP (answer set programming) suffers from a lack of transparency in its outputs. Much like other popular artificial intelligence systems such as deep learning, the results do not come with any explanation to support their derivation. In this paper, we use a given answer set as guidance for a simplified top-down procedure of answer set semantics developed by Satoh and Iwayama to provide not only an explanation for the derivation (or non-derivation) of the atoms, but also an explanation for the consistency of the whole answer set itself. Additionally, we show that a full use of the Satoh-Iwayama procedure gives an explanation of why an atom is not present in any answer set.