Multi-policy optimization in decentralized autonomic systems (extended abstract).
Citation:
Ivana Dusparic and Vinny Cahill., Multi-policy optimization in decentralized autonomic systems (extended abstract)., Proceedings 8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2009), 8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2009), Carles Sierra, Cristiano Castelfranchi, Keith S. Decker, and Jaime Simao Sichman, 2, 2009, 1203-1204Download Item:
Multi-Policy Optimization.pdf (published (publisher copy) peer-reviewed) 141.5Kb
Abstract:
This paper addresses the challenge of multi-policy optimization
in decentralized autonomic systems. We evaluate several
multi-policy reinforcement learning-based optimization
techniques in an urban traffic control simulation, a canonical
example of a decentralized autonomic system. Our results
indicate that W-learning, which learns separately for each
policy and then selects between nominated actions based
on current action importance, is a suitable approach for
optimization towards multiple policies on non-collaborating
agents in heterogeneous autonomic environments.
Author's Homepage:
http://people.tcd.ie/vjcahillhttp://people.tcd.ie/duspari
Description:
PUBLISHED
Author: DUSPARIC, IVANA; CAHILL, VINNY
Other Titles:
Proceedings 8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2009)8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2009)
Type of material:
Conference PaperCollections:
Series/Report no:
2Availability:
Full text availableKeywords:
Computer ScienceLicences: