Robust optimal strategies in Markov decision problems

Gal Oren; Eilon Solan

doi:https://doi.org/10.1016/j.orl.2014.01.001

Robust optimal strategies in Markov decision problems

Gal Oren, Eilon Solan

School of Mathematical Sciences

Tel Aviv University

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

Abstract

An optimal strategy in a Markov decision problem is robust if it is optimal in every decision problem (not necessarily stationary) that is close to the original problem. We prove that when the state and action spaces are finite, an optimal strategy is robust if and only if it is the unique optimal strategy.

Original language	English
Pages (from-to)	109-112
Number of pages	4
Journal	Operations Research Letters
Volume	42
Issue number	2
DOIs	https://doi.org/10.1016/j.orl.2014.01.001
State	Published - Mar 2014

Keywords

Markov decision problems
Optimal strategy
Robustness

All Science Journal Classification (ASJC) codes

Software
Management Science and Operations Research
Industrial and Manufacturing Engineering
Applied Mathematics

Access to Document

https://doi.org/10.1016/j.orl.2014.01.001

Cite this

@article{7c9972a8b2ac4edd9444bcb859912709,

title = "Robust optimal strategies in Markov decision problems",

abstract = "An optimal strategy in a Markov decision problem is robust if it is optimal in every decision problem (not necessarily stationary) that is close to the original problem. We prove that when the state and action spaces are finite, an optimal strategy is robust if and only if it is the unique optimal strategy.",

keywords = "Markov decision problems, Optimal strategy, Robustness",

author = "Gal Oren and Eilon Solan",

note = "Funding Information: Solan acknowledges the support of the Israel Science Foundation , Grant #212/09 and of the Google Inter-university center for Electronic Markets and Auctions .",

year = "2014",

month = mar,

doi = "https://doi.org/10.1016/j.orl.2014.01.001",

language = "الإنجليزيّة",

volume = "42",

pages = "109--112",

journal = "Operations Research Letters",

issn = "0167-6377",

publisher = "Elsevier B.V.",

number = "2",

}

TY - JOUR

T1 - Robust optimal strategies in Markov decision problems

AU - Oren, Gal

AU - Solan, Eilon

N1 - Funding Information: Solan acknowledges the support of the Israel Science Foundation , Grant #212/09 and of the Google Inter-university center for Electronic Markets and Auctions .

PY - 2014/3

Y1 - 2014/3

N2 - An optimal strategy in a Markov decision problem is robust if it is optimal in every decision problem (not necessarily stationary) that is close to the original problem. We prove that when the state and action spaces are finite, an optimal strategy is robust if and only if it is the unique optimal strategy.

AB - An optimal strategy in a Markov decision problem is robust if it is optimal in every decision problem (not necessarily stationary) that is close to the original problem. We prove that when the state and action spaces are finite, an optimal strategy is robust if and only if it is the unique optimal strategy.

KW - Markov decision problems

KW - Optimal strategy

KW - Robustness

UR - http://www.scopus.com/inward/record.url?scp=84893168460&partnerID=8YFLogxK

U2 - https://doi.org/10.1016/j.orl.2014.01.001

DO - https://doi.org/10.1016/j.orl.2014.01.001

M3 - مقالة

SN - 0167-6377

VL - 42

SP - 109

EP - 112

JO - Operations Research Letters

JF - Operations Research Letters

IS - 2

ER -

Robust optimal strategies in Markov decision problems

Abstract

Keywords

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this