2019/03/19 by Sarath Sreedharan, Siddharth Srivastava, Sreedharan, Sarath +6
Computer Science · #AI-based Problem Solving and Planning #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Logic, Reasoning, and Knowledge #Semantic Web and Ontologies
paper · pdf · doi:10.48550/arxiv.1903.08218
openalex publication_date 2019/03/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Explainable planning is widely accepted as a prerequisite for autonomous\nagents to successfully work with humans. While there has been a lot of research\non generating explanations of solutions to planning problems, explaining the\nabsence of solutions remains an open and under-studied problem, even though\nsuch situations can be the hardest to understand or debug. In this paper, we\nshow that hierarchical abstractions can be used to efficiently generate reasons\nfor unsolvability of planning problems. In contrast to related work on\ncomputing certificates of unsolvability, we show that these methods can\ngenerate compact, human-understandable reasons for unsolvability. Empirical\nanalysis and user studies show the validity of our methods as well as their\ncomputational efficacy on a number of benchmark planning domains.\n