Title: Autonomous Network Management in Multi Objective Environments by means of Reinforced Learning Agents
Author: Ricardo Bagnasco
e-mail: rbagnasco@tsc.upc.edu
Partner: UPC
Supervisor: Joan Serrat
Committee:
Year of start: 2007
Year of end: 2011
Funding institution: Ministerio de Innovación y Ciencia (FPI)

This research is intended to contribute to the autonomous management of networks, allowing policies to be dynamically adjusted and aligned to application directives according to the available resources. Many existing management approaches requires static a priori policy deployment but our proposal goes one step further modifying initially deployed policies by learning from the system behaviour. We use a hierarchical policy model to show the connection of high level goals with network level configurations. We also intend to solve two important and mostly forgotten issues: the system has multiple goals some of them contradictory and we will show how to overcome it; and, some current works optimize one network element but being unaware of others participants; instead, our proposed scheme takes into account various social behaviours, such as cooperation and competition among different elements.

1. R. Bagnasco, J. Serrat, “MARL (multi agent reinforcement learning) in P2P Network Management”. 2nd EMANICS Workshop on Peer-to-Peer Management, London, 27-28 April 2009
2. R. Bagnasco, J. Serrat. “Learning and adapting policies in Computer Networks”. ICAC 2009 Doctoral Consortium. 15-19 June 2009 (submitted)
3. R. Bagnasco y J. Serrat. “Multi-Agent Reinforcement Learning in Network Management”, 3th Aims Conference. University of Twente, Netherlands. 30 June, 2009. (submitted)