On Approximate Solutions for Partially Observable Decision Problems

Juri Hinz
Additional contact information
Juri Hinz: University of Technology Sydney

No 421, Research Paper Series from Quantitative Finance Research Centre, University of Technology, Sydney

Abstract: Decision problems under partial observation serve a rich framework for planning and control of operations in diverse applications. In this area, approximate point-based algorithms have emerged improving the scalability of existing numerical schemes to a key level where they can be broadly used in robotics. We elaborate on some core principles of point-based methods and present an approximate solution technique which utilizes linear state dynamics and convexity.

Keywords: Markov decisions; approximate dynamic programming; planning under uncertainty; partially observable Markov decision processes (search for similar items in EconPapers)
Pages: 30 pages
Date: 2021-01-01
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.uts.edu.au/sites/default/files/article/downloads/rp421.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:uts:rpaper:421

Access Statistics for this paper

More papers in Research Paper Series from Quantitative Finance Research Centre, University of Technology, Sydney PO Box 123, Broadway, NSW 2007, Australia. Contact information at EDIRC.
Bibliographic data for series maintained by Duncan Ford ().