Paper published in a book (Scientific congresses and symposiums)
Using prior knowledge to accelerate online least-squares policy iteration
Busoniu, Lucian; De Schutter, Bart; Babuska, Robert et al.
2010In Proceedings of the 2010 IEEE International Conference on Automation, Quality and Testing, Robotics
Peer reviewed
 

Files


Full Text
onlineLSPI-priorknowledge2010.pdf
Publisher postprint (957.05 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Reinforcement learning; on-line LSPI; prior knowledge
Abstract :
[en] Reinforcement learning (RL) is a promising paradigm for learning optimal control. Although RL is generally envisioned as working without any prior knowledge about the system, such knowledge is often available and can be exploited to great advantage. In this paper, we consider prior knowledge about the monotonicity of the control policy with respect to the system states, and we introduce an approach that exploits this type of prior knowledge to accelerate a state-of-the-art RL algorithm called online least-squares policy iteration (LSPI). Monotonic policies are appropriate for important classes of systems appearing in control applications. LSPI is a data-efficient RL algorithm that we previously extended to online learning, but that did not provide until now a way to use prior knowledge about the policy. In an empirical evaluation, online LSPI with prior knowledge learns much faster and more reliably than the original online LSPI.
Disciplines :
Computer science
Author, co-author :
Busoniu, Lucian
De Schutter, Bart
Babuska, Robert
Ernst, Damien  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation
Language :
English
Title :
Using prior knowledge to accelerate online least-squares policy iteration
Publication date :
May 2010
Event name :
2010 IEEE International Conference on Automation, Quality and Testing, Robotics (AQTR 2010)
Event place :
Cluj-Napoca, Romania
Event date :
28-30 May 2010
Audience :
International
Main work title :
Proceedings of the 2010 IEEE International Conference on Automation, Quality and Testing, Robotics
ISBN/EAN :
978-1-4244-6724-2
Peer reviewed :
Peer reviewed
Available on ORBi :
since 02 June 2010

Statistics


Number of views
56 (3 by ULiège)
Number of downloads
204 (2 by ULiège)

Scopus citations®
 
8
Scopus citations®
without self-citations
7
OpenCitations
 
5

Bibliography


Similar publications



Contact ORBi