Epistemic Exploration for Generalizable Planning and Learning in Non-Stationary Settings

Rushang Karia, Pulkit Verma, Alberto Speranzon, Siddharth Srivastava

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

1 Scopus citations

Abstract

This paper introduces a new approach for continual planning and model learning in relational, non-stationary stochastic environments. Such capabilities are essential for the deployment of sequential decision-making systems in the uncertain and constantly evolving real world. Working in such practical settings with unknown (and non-stationary) transition systems and changing tasks, the proposed framework models gaps in the agent's current state of knowledge and uses them to conduct focused, investigative explorations. Data collected using these explorations is used for learning generalizable probabilistic models for solving the current task despite continual changes in the environment dynamics. Empirical evaluations on several non-stationary benchmark domains show that this approach significantly outperforms planning and RL baselines in terms of sample complexity. Theoretical results show that the system exhibits desirable convergence properties when stationarity holds.

Original languageEnglish (US)
Title of host publicationProceedings of the 34th International Conference on Automated Planning and Scheduling, ICAPS 2024
EditorsSara Bernardini, Christian Muise
PublisherAssociation for the Advancement of Artificial Intelligence
Pages310-318
Number of pages9
ISBN (Electronic)9781577358893
DOIs
StatePublished - May 30 2024
Event34th International Conference on Automated Planning and Scheduling, ICAPS 2024 - Banaff, Canada
Duration: Jun 1 2024Jun 6 2024

Publication series

NameProceedings International Conference on Automated Planning and Scheduling, ICAPS
Volume34
ISSN (Print)2334-0835
ISSN (Electronic)2334-0843

Conference

Conference34th International Conference on Automated Planning and Scheduling, ICAPS 2024
Country/TerritoryCanada
CityBanaff
Period6/1/246/6/24

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Information Systems and Management

Fingerprint

Dive into the research topics of 'Epistemic Exploration for Generalizable Planning and Learning in Non-Stationary Settings'. Together they form a unique fingerprint.

Cite this