Application of Newton's method to action selection in continuous state- and action-space reinforcement learning

Nichols, Barry D. and Dracopoulos, Dimitris C. (2014) Application of Newton's method to action selection in continuous state- and action-space reinforcement learning. In: ESANN 2014 Proceedings: 22nd European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. Bruges April 23-24-25, 2014. ESANN, pp. 141-146. ISBN 9782874190957

[img]
Preview
PDF - Published version (with publisher's formatting)
Download (1MB) | Preview

Abstract

An algorithm based on Newton’s Method is proposed for action selection in continuous state- and action-space reinforcement learning without a policy network or discretization. The proposed method is validated on two benchmark problems: Cart-Pole and double Cart-Pole on which the proposed method achieves comparable or improved performance with less parameters to tune and in less training episodes than CACLA, which has previously been shown to outperform many other continuous state- and action-space reinforcement learning algorithms.

Item Type: Book Section
Research Areas: A. > School of Science and Technology > Computer Science
Item ID: 17064
Useful Links:
Depositing User: Barry Nichols
Date Deposited: 24 Jun 2015 14:57
Last Modified: 07 Dec 2018 08:34
URI: http://eprints.mdx.ac.uk/id/eprint/17064

Actions (login required)

Edit Item Edit Item

Full text downloads (NB count will be zero if no full text documents are attached to the record)

Downloads per month over the past year