Planning with Partially Observable Markov Decision Processes:
Advances in Exact Solution Method
Nevin L. Zhang and Stephen S. Lee:
N. L. Zhang
Hong Kong University of Science and Technology
Clear Water Bay Road
Kowloon, Hong Kong
e-mail: EMAIL ADDRESS
Phone: (+852)-2358-7015
FAX : (+852)-2358-1477
S. S. Lee
Hong Kong University of Science and Technology
Clear Water Bay Road
Kowloon, Hong Kong
e-mail: EMAIL ADDRESS
Abstract:
There is much interest in using partially observable Markov
decision processes (POMDPs) as a formal model for
planning in stochastic domains. This paper
is concerned with finding optimal policies for
POMDPs. We propose several improvements to
incremental pruning, presently the most efficient
exact algorithm for solving POMDPs.
Keywords:
Planning under uncertainty,
Partially observable Markov decision processes,
exact algorithms, incremental pruning.
Availability:
This paper is also available in
PostScript
format.