Fast Value Iteration for Goal-Directed Markov Decision Processes
Artificial Intelligence
2013-02-08 v1
Abstract
Planning problems where effects of actions are non-deterministic can be modeled as Markov decision processes. Planning problems are usually goal-directed. This paper proposes several techniques for exploiting the goal-directedness to accelerate value iteration, a standard algorithm for solving Markov decision processes. Empirical studies have shown that the techniques can bring about significant speedups.
Cite
@article{arxiv.1302.1575,
title = {Fast Value Iteration for Goal-Directed Markov Decision Processes},
author = {Nevin Lianwen Zhang and Weihong Zhang},
journal= {arXiv preprint arXiv:1302.1575},
year = {2013}
}
Comments
Appears in Proceedings of the Thirteenth Conference on Uncertainty in Artificial Intelligence (UAI1997)