IEEE - Institute of Electrical and Electronics Engineers, Inc. - Constructing action set from basis functions for reinforcement learning of robot control

2009 IEEE International Conference on Robotics and Automation (ICRA)

Author(s): Yamaguchi, Akihiko ; Jun Takamatsu, ; Tsukasa Ogasawara,
Publisher: IEEE - Institute of Electrical and Electronics Engineers, Inc.
Publication Date: 1 May 2009
Conference Location: Kobe, Japan, Japan
Conference Date: 12 May 2009
Page(s): 2,525 - 2,532
ISBN (CD): 978-1-4244-2789-5
ISBN (Paper): 978-1-4244-2788-8
ISSN (Paper): 1050-4729
DOI: 10.1109/ROBOT.2009.5152840
Regular:

Continuous action sets are used in many reinforcement learning (RL) applications in robot control since the control input is continuous. However, discrete action sets also have the advantages of... View More

Advertisement