Variable Resolution Dynamic Programming: Efficiently Learning Action Maps in Multivariate Real-valued State-spaces
Conference Paper, Proceedings of (ICML) International Conference on Machine Learning, pp. 333 - 337, June, 1991
Abstract
An effective method to create an autonomous reactive controller is to learn a model of the environment and then use dynamic programming to derive a policy to maximize long term reward. Neither learning environmental models nor dynamic programming require parametric assumptions about the world, and so learning can proceed with no danger of becoming “stuck― by a mismatch between the parametric assumptions and reality. The paper discusses how such an approach can be realized in real valued multivariate state spaces in which straightforward discretization falls prey to the curse of dimensionality.
BibTeX
@conference{Moore-1991-15826,author = {Andrew Moore},
title = {Variable Resolution Dynamic Programming: Efficiently Learning Action Maps in Multivariate Real-valued State-spaces},
booktitle = {Proceedings of (ICML) International Conference on Machine Learning},
year = {1991},
month = {June},
editor = {L. Birnbaum and G. Collins},
pages = {333 - 337},
publisher = {Morgan Kaufmann},
}
Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.