English

Object-oriented state editing for HRL

Machine Learning 2019-11-01 v1 Artificial Intelligence Machine Learning

Abstract

We introduce agents that use object-oriented reasoning to consider alternate states of the world in order to more quickly find solutions to problems. Specifically, a hierarchical controller directs a low-level agent to behave as if objects in the scene were added, deleted, or modified. The actions taken by the controller are defined over a graph-based representation of the scene, with actions corresponding to adding, deleting, or editing the nodes of a graph. We present preliminary results on three environments, demonstrating that our approach can achieve similar levels of reward as non-hierarchical agents, but with better data efficiency.

Keywords

Cite

@article{arxiv.1910.14361,
  title  = {Object-oriented state editing for HRL},
  author = {Victor Bapst and Alvaro Sanchez-Gonzalez and Omar Shams and Kimberly Stachenfeld and Peter W. Battaglia and Satinder Singh and Jessica B. Hamrick},
  journal= {arXiv preprint arXiv:1910.14361},
  year   = {2019}
}

Comments

8 pages; accepted to the Perception as Generative Reasoning workshop of the 33rd Conference on Neural InformationProcessing Systems (NeurIPS 2019)

R2 v1 2026-06-23T12:00:36.455Z