Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

2-2017

Abstract

We propose a model-based approach to hierarchical reinforcement learning that exploits shared knowledge and selective execution at different levels of abstraction, to efficiently solve large, complex problems. Our framework adopts a new transition dynamics learning algorithm that identifies the common action-feature combinations of the subtasks, and evaluates the subtask execution choices through simulation. The framework is sample efficient, and tolerates uncertain and incomplete problem characterization of the subtasks. We test the framework on common benchmark problems and complex simulated robotic environments. It compares favorably against the stateof-the-art algorithms, and scales well in very large problems.

Keywords

Reinforcement learning, hierarchical reinforcement learning, MAXQ, R-MAX, model-based reinforcement learning, Bench-mark problems, Feature combination, Hierarchical reinforcement learning, Levels of abstraction, Model based approach, Problem characterization, Robotic environments, State-of-the-art algorithms

Discipline

Artificial Intelligence and Robotics | Operations Research, Systems Engineering and Industrial Engineering | Theory and Algorithms

Research Areas

Intelligent Systems and Optimization

Publication

Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17): San Francisco, CA, February 4-10

First Page

3583

Last Page

3589

Publisher

AAAI Press

City or Country

Menlo Park, CA

Copyright Owner and License

Publisher

Additional URL

https://aaai.org/ocs/index.php/AAAI/AAAI17/paper/view/14771

Download

Share

COinS