Discovering User-Interpretable Capabilities of Black-Box Planning Agents

Pulkit Verma, Shashank Rao Marpally, Siddharth Srivastava

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Several approaches have been developed for answering users' specific questions about AI behavior and for assessing their core functionality in terms of primitive executable actions. However, the problem of summarizing an AI agent's broad capabilities for a user is comparatively new. This paper presents an algorithm for discovering from scratch the suite of high-level “capabilities” that an AI system with arbitrary internal planning algorithms/policies can perform. It computes conditions describing the applicability and effects of these capabilities in user-interpretable terms. Starting from a set of user-interpretable state properties, an AI agent, and a simulator that the agent can interact with, our algorithm returns a set of high-level capabilities with their parameterized descriptions. Empirical evaluation on several game-based scenarios shows that this approach efficiently learns descriptions of various types of AI agents in deterministic, fully observable settings. User studies show that such descriptions are easier to understand and reason with than the agent's primitive actions.

Original languageEnglish (US)
Title of host publication19th International Conference on Principles of Knowledge Representation and Reasoning, KR 2022
PublisherInternational Joint Conferences on Artificial Intelligence
Pages362-372
Number of pages11
ISBN (Electronic)9781956792010
StatePublished - 2022
Event19th International Conference on Principles of Knowledge Representation and Reasoning, KR 2022 - Haifa, Israel
Duration: Jul 31 2022Aug 5 2022

Publication series

Name19th International Conference on Principles of Knowledge Representation and Reasoning, KR 2022

Conference

Conference19th International Conference on Principles of Knowledge Representation and Reasoning, KR 2022
Country/TerritoryIsrael
CityHaifa
Period7/31/228/5/22

ASJC Scopus subject areas

  • Software
  • Logic

Fingerprint

Dive into the research topics of 'Discovering User-Interpretable Capabilities of Black-Box Planning Agents'. Together they form a unique fingerprint.

Cite this