Reactive Control of Dynamic Progressive Processing

Shlomo Zilberstein and Abdel-Illah Mouaddib. Reactive Control of Dynamic Progressive Processing. Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence (IJCAI), 1268-1273, Stockholm, Sweden, 1999.

Abstract

Progressive processing is a model of computation that allows a system to tradeoff computational resources against the quality of results. This paper generalizes the existing model to make it suitable for dynamic composition of information retrieval techniques. The new framework addresses effectively the uncertainty associated with the duration and output quality of each component. We show how to construct an optimal meta-level controller for a single task based on solving a corresponding Markov decision problem, and how to extend the solution to the case of multiple and dynamic tasks using the notion of an opportunity cost.

Bibtex entry:

@inproceedings{ZMijcai99,
  author	= {Shlomo Zilberstein and Abdel-Illah Mouaddib},
  title		= {Reactive Control of Dynamic Progressive Processing},
  booktitle     = {Proceedings of the Sixteenth International Joint Conference on
		   Artificial Intelligence},
  year		= {1999},
  pages		= {1268-1273},
  address       = {Stockholm, Sweden},
  url		= {http://rbr.cs.umass.edu/shlomo/papers/ZMijcai99.html}
}

shlomo@cs.umass.edu
UMass Amherst