Global Policy Construction in Modular Reinforcement Learning

Authors

  • Ruohan Zhang The University of Texas at Austin
  • Zhao Song The University of Texas at Austin
  • Dana Ballard The University of Texas at Austin

DOI:

https://doi.org/10.1609/aaai.v29i1.9736

Keywords:

Modular reinforcemenet learning

Abstract

We propose a modular reinforcement learning algorithm which decomposes a Markov decision process into independent modules. Each module is trained using Sarsa(lambda). We introduce three algorithms for forming global policy from modules policies, and demonstrate our results using a 2D grid world.

AAAI 2015 Proceedings Cover

Downloads

Published

2015-03-04

How to Cite

Zhang, R., Song, Z., & Ballard, D. (2015). Global Policy Construction in Modular Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 29(1). https://doi.org/10.1609/aaai.v29i1.9736

Issue

Vol. 29 No. 1 (2015): Twenty-Ninth AAAI Conference on Artificial Intelligence

Section

Student Abstract Track