cs.LGFeb 2, 2026

Joint Learning of Hierarchical Neural Options and Abstract World Model

Authors: Wasu Top Piriyakulkij, Wolfgang Lehrach, Kevin Ellis, Kevin Murphy

Organizations: Cornell University · Google Deepmind

Abstract

Building agents that can perform new skills by composing existing skills is a long-standing goal of AI agent research. Towards this end, we investigate how to efficiently acquire a sequence of skills, formalized as hierarchical neural options. However, existing model-free hierarchical reinforcement algorithms need a lot of data. We propose a novel method, which we call AgentOWL (Option and World model Learning Agent), that jointly learns -- in a sample efficient way -- an abstract world model (abstracting across both states and time) and a set of hierarchical neural options. We show, on a subset of Object-Centric Atari games, that our method can learn more skills using less data than baseline methods and possesses learning and generalization capabilities that the baselines do not have.

Figures & tables

Appendix figures & tables12 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration for ARC-AGI-3

    Jul 1, 2026David Courtis, Wenhao Li, Scott SannerWorld Model LearningWorld Models

  2. How do World Models and Policies Compose in LLM Agents? A Joint Spectral and Behavioral Account

    Aug 30, 2026Ruize Xu, Xiao Yu, Yujin Tang +2World Model LearningWorld Models