2020/04/16 by Mehrdad Zakershahrak, Zakershahrak, Mehrdad, Shashank Rao Marpally +7 · 4 citations
Computer Science · Mathematics · Psychology · #AI-based Problem Solving and Planning #Artificial Intelligence (cs.AI) #Artificial intelligence #Cognitive science #Computer science #Context (archaeology) #Domain (mathematical analysis) #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Function (biology) #Human–computer interaction #Independence (probability theory) #Mathematics #Order (exchange) #Perspective (graphical) #Preference #Process (computing) #Psychology #Reinforcement Learning in Robotics #Reinforcement learning #Set (abstract data type) #cs.AI
paper · pdf · doi:10.48550/arxiv.2004.07822
published in arXiv (Cornell University) (Cornell University) · arXiv admin note: text overlap with arXiv:1902.00604
openalex publication_date 2020/04/16 · openalex created_date 2020/04/24 · arxiv created 2020/10/17 · arxiv updated 2020/10/20 · openalex updated_date 2026/08/06
Prior work on generating explanations in a planning and decision-making context has focused on providing the rationale behind an AI agent's decision making. While these methods provide the right explanations from the explainer's perspective, they fail to heed the cognitive requirement of understanding an explanation from the explainee's (the human's) perspective. In this work, we set out to address this issue by first considering the influence of information order in an explanation, or the progressiveness of explanations. Intuitively, progression builds later concepts on previous ones and is known to contribute to better learning. In this work, we aim to investigate similar effects during explanation generation when an explanation is broken into multiple parts that are communicated sequentially. The challenge here lies in modeling the humans' preferences for information order in receiving such explanations to assist understanding. Given this sequential process, a formulation based on goal-based MDP for generating progressive explanations is presented. The reward function of this MDP is learned via inverse reinforcement learning based on explanations that are retrieved via human subject studies. We first evaluated our approach on a scavenger-hunt domain to demonstrate its effectively in capturing the humans' preferences. Upon analyzing the results, it revealed something more fundamental: the preferences arise strongly from both domain dependent and independence features. The correlation with domain independent features pushed us to verify this result further in an escape room domain. Results confirmed our hypothesis that the process of understanding an explanation was a dynamic process. The human preference that reflected this aspect corresponded exactly to the progression for knowledge assimilation hidden deeper in our cognitive process.