Distributed reinforcement learning of autonomic behaviour.
Item Type:Conference Paper
Citation:Jim Dowling, Raymond Cunningham, Eoin Curran, and Vinny Cahill., Distributed reinforcement learning of autonomic behaviour., 2nd International Workshop on Self-Adaptive and Autonomic Computing Systems (SAACS'04), Zaragoza, Spain, September, 2004
Collaborative.pdf (published (author copy) peer-reviewed) 80.50Kb
This paper introduces Collaborative Reinforcement Learning (CRL), a coordination model for solving system-wide optimisation problems in distributed systems where there is no support for global state. In CRL the autonomic properties of a distributed system emerge from the coordination of individual agents solving discrete optimisation problems using Reinforcement Learning. In the context of an ad hoc routing protocol, we show how system-wide optimisation in CRL can be used to establish and maintain autonomic properties for decentralised distributed systems.
Winner Best Paper Award.
Author: CAHILL, VINNY
Other Titles:2nd International Workshop on Self-Adaptive and Autonomic Computing Systems (SAACS'04)
Type of material:Conference Paper
Availability:Full text available