A note on policy algorithms for discounted Markov decision problems

Michael K. Ng

doi:10.1016/S0167-6377(99)00051-6

A note on policy algorithms for discounted Markov decision problems

Michael K. Ng^*

^*Corresponding author for this work

Department of Mathematics

Research output: Contribution to journal › Journal article › peer-review

4 Citations (Scopus)

Abstract

The evaluation phase in the policy iteration algorithm for the infinite horizon discounted Markov decision problem is presented, which can be done in O(mN²) operations, where N is the number of states of the Markov decision process and m is the number of states in which the decision changes during the policy improvement phase.

Original language	English
Pages (from-to)	195-197
Number of pages	3
Journal	Operations Research Letters
Volume	25
Issue number	4
DOIs	https://doi.org/10.1016/S0167-6377(99)00051-6
Publication status	Published - Nov 1999

User-Defined Keywords

Discounted Markov decision process
Policy algorithm
Matrices

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/S0167-6377(99)00051-6

Cite this

@article{bb2f1a53a5c948bfa4e3e3eab2abe022,

title = "A note on policy algorithms for discounted Markov decision problems",

abstract = "The evaluation phase in the policy iteration algorithm for the infinite horizon discounted Markov decision problem is presented, which can be done in O(mN2) operations, where N is the number of states of the Markov decision process and m is the number of states in which the decision changes during the policy improvement phase.",

keywords = "Discounted Markov decision process, Policy algorithm, Matrices",

author = "Ng, {Michael K.}",

note = "Research supported by HKU CRCG grant nos. 10201939 and 10201824. Publisher Copyright: {\textcopyright} 1999 Elsevier Science B.V. ",

year = "1999",

month = nov,

doi = "10.1016/S0167-6377(99)00051-6",

language = "English",

volume = "25",

pages = "195--197",

journal = "Operations Research Letters",

issn = "0167-6377",

publisher = "Elsevier B.V.",

number = "4",

}

TY - JOUR

T1 - A note on policy algorithms for discounted Markov decision problems

AU - Ng, Michael K.

PY - 1999/11

Y1 - 1999/11

N2 - The evaluation phase in the policy iteration algorithm for the infinite horizon discounted Markov decision problem is presented, which can be done in O(mN2) operations, where N is the number of states of the Markov decision process and m is the number of states in which the decision changes during the policy improvement phase.

AB - The evaluation phase in the policy iteration algorithm for the infinite horizon discounted Markov decision problem is presented, which can be done in O(mN2) operations, where N is the number of states of the Markov decision process and m is the number of states in which the decision changes during the policy improvement phase.

KW - Discounted Markov decision process

KW - Policy algorithm

KW - Matrices

UR - http://www.scopus.com/inward/record.url?scp=0033356671&partnerID=8YFLogxK

UR - https://www.sciencedirect.com/science/article/pii/S0167637799000516?via%3Dihub

U2 - 10.1016/S0167-6377(99)00051-6

DO - 10.1016/S0167-6377(99)00051-6

M3 - Journal article

AN - SCOPUS:0033356671

SN - 0167-6377

VL - 25

SP - 195

EP - 197

JO - Operations Research Letters

JF - Operations Research Letters

IS - 4

ER -

A note on policy algorithms for discounted Markov decision problems

Abstract

User-Defined Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this