Publication - Multiagent Meta-level Control for Predicting Meteorological Phenomena

Authors: Cheng, Shanjun; Raja, Anita; Lesser, Victor
Title: Multiagent Meta-level Control for Predicting Meteorological Phenomena
Abstract: It is crucial for embedded systems to adapt to the dynamics of open environments. This adaptation process becomes especially challenging in the context of multiagent systems. In this paper, we argue that multiagent meta-level control is an effective way to determine when this adaptation process should be done and how much effort should be invested in adaptation as opposed to continuing with the current action plan. We develop a reinforcement learning-based mechanism for multiagent meta-level control that facilitates the metalevel control component of each agent to learn policies in a decentralized fashion that (a) it can efficiently support agent interactions with other agents and (b) reorganize the underlying network when needed. We evaluate this mechanism in the context of a multiagent tornado tracking application called NetRads. Empirical results show that adaptive multiagent meta-level control significantly improves the performance of the tornado tracking network for a variety of weather scenarios.
Publication: Proceedings of IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT-10), pp. 157 - 164
Location: Toronto
Publisher: IEEE Computer Society
Date: August 2010
Sources: PDF: /Documents/lesser/Cheng_IAT10.pdf
Reference: Cheng, Shanjun; Raja, Anita; Lesser, Victor. Multiagent Meta-level Control for Predicting Meteorological Phenomena. Proceedings of IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT-10), IEEE Computer Society, pp. 157-164. August 2010.
bibtex: 
@inproceedings{Cheng-490,
  author    = "Shanjun Cheng and Anita Raja and Victor Lesser",
  title     = "{Multiagent Meta-level Control for Predicting
               Meteorological Phenomena}",
  booktitle = "Proceedings of  IEEE/WIC/ACM International
               Conference on Intelligent Agent Technology
               (IAT-10)",
  publisher = "IEEE Computer Society",
  pages     = "157-164",
  month     = "August",
  year      = "2010",
  address   = "Toronto",
  url       = "http://mas.cs.umass.edu/paper/490",
}