When Agents Talk Back: Rebellious ExplanationsDownload PDF

Anonymous

Published: 24 May 2019, Last Modified: 05 May 2023XAIP 2019Readers: Everyone
Keywords: Rebellion, Deception, Explanation
TL;DR: Position paper proposing rebellious and deceptive explanations for agents.
Abstract: As the area of Explainable AI (XAI), and Explainable AI Planning (XAIP), matures, the ability for agents to generate and curate explanations will likewise grow. We propose a new challenge area in the form of rebellious and deceptive explanations. We discuss how these explanations might be generated and then briefly discuss evaluation criteria.
5 Replies

Loading