On constrained Markov decision processes

Moshe Haviv

doi:10.1016/0167-6377(96)00003-X

On constrained Markov decision processes

Moshe Haviv^*

^*Corresponding author for this work

Department of Statistics

Research output: Contribution to journal › Article › peer-review

29 Scopus citations

Abstract

A multichain Markov decision process with constraints on the expected state-action frequencies may lead to a unique optimal policy which does not satisfy Bellman's principle of optimality. The model with sample-path constraints does not suffer from this drawback.

Original language	English
Pages (from-to)	25-28
Number of pages	4
Journal	Operations Research Letters
Volume	19
Issue number	1
DOIs	https://doi.org/10.1016/0167-6377(96)00003-X
State	Published - Jul 1996

Keywords

Constrained optimization
Markov processes
Sample path

Access to Document

10.1016/0167-6377(96)00003-X

Cite this

@article{468bc40dbe3349658f665c0474838bb6,

title = "On constrained Markov decision processes",

abstract = "A multichain Markov decision process with constraints on the expected state-action frequencies may lead to a unique optimal policy which does not satisfy Bellman's principle of optimality. The model with sample-path constraints does not suffer from this drawback.",

keywords = "Constrained optimization, Markov processes, Sample path",

author = "Moshe Haviv",

year = "1996",

month = jul,

doi = "10.1016/0167-6377(96)00003-X",

language = "אנגלית",

volume = "19",

pages = "25--28",

journal = "Operations Research Letters",

issn = "0167-6377",

publisher = "Elsevier B.V.",

number = "1",

}

TY - JOUR

T1 - On constrained Markov decision processes

AU - Haviv, Moshe

PY - 1996/7

Y1 - 1996/7

N2 - A multichain Markov decision process with constraints on the expected state-action frequencies may lead to a unique optimal policy which does not satisfy Bellman's principle of optimality. The model with sample-path constraints does not suffer from this drawback.

AB - A multichain Markov decision process with constraints on the expected state-action frequencies may lead to a unique optimal policy which does not satisfy Bellman's principle of optimality. The model with sample-path constraints does not suffer from this drawback.

KW - Constrained optimization

KW - Markov processes

KW - Sample path

UR - https://www.scopus.com/pages/publications/0009421908

U2 - 10.1016/0167-6377(96)00003-X

DO - 10.1016/0167-6377(96)00003-X

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:0009421908

SN - 0167-6377

VL - 19

SP - 25

EP - 28

JO - Operations Research Letters

JF - Operations Research Letters

IS - 1

ER -

On constrained Markov decision processes

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this