Monotonically improving limit-optimal strategies in finite-state decision processes

T.P. Hill, J. Wal, van der

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

In every finite-state leavable gambling problem and in every finite-state Markov decision process with discounted, negative or positive reward criteria there exists a Markov strategy which is monotonically improving and optimal in the limit along every history. An example is given to show that for the positive and gambling cases such strategies cannot be constructed by simply switching to a "better" action or gamble at each successive return to a state.
Original languageEnglish
Pages (from-to)463-473
Number of pages11
JournalMathematics of Operations Research
Volume12
Issue number3
DOIs
Publication statusPublished - 1987

Fingerprint

Gambling
Optimal Strategy
Gamble
Markov Decision Process
Reward
Strategy
Optimal strategy
Decision process
History
Gambles
Markov decision process
Markov strategies

Cite this

@article{703f250ec4af44a290ac3753ffc90bb8,
title = "Monotonically improving limit-optimal strategies in finite-state decision processes",
abstract = "In every finite-state leavable gambling problem and in every finite-state Markov decision process with discounted, negative or positive reward criteria there exists a Markov strategy which is monotonically improving and optimal in the limit along every history. An example is given to show that for the positive and gambling cases such strategies cannot be constructed by simply switching to a {"}better{"} action or gamble at each successive return to a state.",
author = "T.P. Hill and {Wal, van der}, J.",
year = "1987",
doi = "10.1287/moor.12.3.463",
language = "English",
volume = "12",
pages = "463--473",
journal = "Mathematics of Operations Research",
issn = "0364-765X",
publisher = "INFORMS Institute for Operations Research and the Management Sciences",
number = "3",

}

Monotonically improving limit-optimal strategies in finite-state decision processes. / Hill, T.P.; Wal, van der, J.

In: Mathematics of Operations Research, Vol. 12, No. 3, 1987, p. 463-473.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Monotonically improving limit-optimal strategies in finite-state decision processes

AU - Hill, T.P.

AU - Wal, van der, J.

PY - 1987

Y1 - 1987

N2 - In every finite-state leavable gambling problem and in every finite-state Markov decision process with discounted, negative or positive reward criteria there exists a Markov strategy which is monotonically improving and optimal in the limit along every history. An example is given to show that for the positive and gambling cases such strategies cannot be constructed by simply switching to a "better" action or gamble at each successive return to a state.

AB - In every finite-state leavable gambling problem and in every finite-state Markov decision process with discounted, negative or positive reward criteria there exists a Markov strategy which is monotonically improving and optimal in the limit along every history. An example is given to show that for the positive and gambling cases such strategies cannot be constructed by simply switching to a "better" action or gamble at each successive return to a state.

UR - http://www.jstor.org/stable/3689977

U2 - 10.1287/moor.12.3.463

DO - 10.1287/moor.12.3.463

M3 - Article

VL - 12

SP - 463

EP - 473

JO - Mathematics of Operations Research

JF - Mathematics of Operations Research

SN - 0364-765X

IS - 3

ER -