Prioritization Methods for Accelerating MDP Solvers
Abstract
The performance of value and policy iteration can be dramatically improved by eliminating redundant or useless backups, and by backing up states in the right order. We study several methods designed to accelerate these iterative solvers, including prioritization, partitioning, and variable reordering. We generate a family of algorithms by combining several of the methods discussed, and present extensive empirical evidence demonstrating that performance can improve by several orders of magnitude for many problems, while preserving accuracy and convergence guarantees.
Cite
Text
Wingate and Seppi. "Prioritization Methods for Accelerating MDP Solvers." Journal of Machine Learning Research, 2005.Markdown
[Wingate and Seppi. "Prioritization Methods for Accelerating MDP Solvers." Journal of Machine Learning Research, 2005.](https://mlanthology.org/jmlr/2005/wingate2005jmlr-prioritization/)BibTeX
@article{wingate2005jmlr-prioritization,
title = {{Prioritization Methods for Accelerating MDP Solvers}},
author = {Wingate, David and Seppi, Kevin D.},
journal = {Journal of Machine Learning Research},
year = {2005},
pages = {851-881},
volume = {6},
url = {https://mlanthology.org/jmlr/2005/wingate2005jmlr-prioritization/}
}