CJO - Abstract - Building autonomic systems using collaborative reinforcement learning

Cambridge Journals Online

Cambridge Journals Online
Skip to content
The Knowledge Engineering Review (2006), 21 : 231-238 Cambridge University Press
Copyright © 2006 Cambridge University Press
doi:10.1017/S0269888906000956 (About doi)
Published online by Cambridge University Press 19 Oct 2006
 
The Knowledge Engineering Review (2006), 21:3:231-238 Cambridge University Press
Copyright © 2006 Cambridge University Press
doi:10.1017/S0269888906000956

Building autonomic systems using collaborative reinforcement learning

JIM  DOWLING a1 , RAYMOND  CUNNINGHAM a1 , EOIN  CURRAN a1 and VINNY  CAHILL a1
a1 Distributed Systems Group, Department of Computer Science, Trinity College, Dublin; e-mail: jdowling@cs.tcd.ie, rcnnnghm@cs.tcd.ie, currane@maths.tcd.ie, vjcahill@cs.tcd.ie

Article author query
dowling j   [Google Scholar] 
cunningham r   [Google Scholar] 
curran e   [Google Scholar] 
cahill v   [Google Scholar] 
 

Abstract

This paper presents Collaborative Reinforcement Learning (CRL), a coordination model for online system optimization in decentralized multi-agent systems. In CRL system optimization problems are represented as a set of discrete optimization problems, each of whose solution cost is minimized by model-based reinforcement learning agents collaborating on their solution. CRL systems can be built to provide autonomic behaviours such as optimizing system performance in an unpredictable environment and adaptation to partial failures. We evaluate CRL using an ad hoc routing protocol that optimizes system routing performance in an unpredictable network environment.


back to top
Cambridge University Press