Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

5-2016

Abstract

We propose a new, core task abstraction (CTA) approach to learning the relevant transition functions in model-based hierarchical reinforcement learning. CTA exploits contextual independences of the state variables conditional on the task-specific actions; its promising performance is demonstrated through a set of benchmark problems.

Keywords

hierarchical reinforcement learning, core task abstraction

Discipline

Computer Sciences | Numerical Analysis and Scientific Computing | Theory and Algorithms

Research Areas

Intelligent Systems and Optimization

Publication

AAMAS '16: Proceedings of the 2016 International Conference on Autonomous Agents and Multiagent Systems: May 9-13, 2016, Singapore

First Page

1411

Last Page

1412

ISBN

9781450342391

Publisher

IFAAMAS

City or Country

Ann Arbor, MI

Copyright Owner and License

Publisher

Additional URL

http://www.ifaamas.org/Proceedings/aamas2016/pdfs/p1411.pdf

Share

COinS