Mental Actions and Explainability in Kripkean Semantics: What Else do I Know? (Student Abstract)

No. 18: AAAI-21 Student Papers and Demonstrations

Volume

Proceedings of the AAAI Conference on Artificial Intelligence, 35

AAAI Student Abstract and Poster Program

The ability of an agent to distinguish the ramification effects of an action from its direct effects adds value to the explainability of its decisions. In this work, we propose to encode the ramification effects of ontic and epistemic actions as single-point update models in an epistemic planning domain modeled with Kripkean semantics of Knowledge and Belief. We call them “mental actions”. We discuss a preliminary approach to realize our idea, and we conclude by pointing out some optimizations as our ongoing pursuit.

10.1609/aaai.v35i18.17943