Abstract
As discussed in Chap. 6, infinite-horizon planning methods focus on generating ε-optimal solutions or solutions with a fixed controller size. We first discuss policy iteration, which is an extension of dynamic programming for Dec-POMDPs to the infinite-horizon case and can produce ε-optimal solutions. We then describe some of the heuristic algorithms that generate fixed-size controllers.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2016 The Author(s)
About this chapter
Cite this chapter
Oliehoek, F.A., Amato, C. (2016). Infinite-Horizon Planning Methods: Discounted Cumulative Reward. In: A Concise Introduction to Decentralized POMDPs. SpringerBriefs in Intelligent Systems. Springer, Cham. https://doi.org/10.1007/978-3-319-28929-8_7
Download citation
DOI: https://doi.org/10.1007/978-3-319-28929-8_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-28927-4
Online ISBN: 978-3-319-28929-8
eBook Packages: Computer ScienceComputer Science (R0)