Rollout Algorithms for Combinatorial Optimization

Dimitri P. Bertsekas; John N. Tsitsiklis; Cynara Wu

doi:10.1023/A:1009635226865

Rollout Algorithms for Combinatorial Optimization

Dimitri P. Bertsekas, John N. Tsitsiklis, Cynara Wu

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

149 Scopus citations

Abstract

We consider the approximate solution of discrete optimization problems using procedures that are capable of magnifying the effectiveness of any given heuristic algorithm through sequential application. In particular, we embed the problem within a dynamic programming framework, and we introduce several types of rollout algorithms, which are related to notions of policy iteration. We provide conditions guaranteeing that the rollout algorithm improves the performance of the original heuristic algorithm. The method is illustrated in the context of a machine maintenance and repair problem.

Original language	English (US)
Pages (from-to)	245-262
Number of pages	18
Journal	Journal of Heuristics
Volume	3
Issue number	3
DOIs	https://doi.org/10.1023/A:1009635226865
State	Published - 1997
Externally published	Yes

ASJC Scopus subject areas

Software
Information Systems
Computer Networks and Communications
Control and Optimization
Management Science and Operations Research
Artificial Intelligence

Access to Document

10.1023/A:1009635226865

Fingerprint Dive into the research topics of 'Rollout Algorithms for Combinatorial Optimization'. Together they form a unique fingerprint.

Cite this

@article{ad57de1f74a24286b4134088f1d8fc8f,

title = "Rollout Algorithms for Combinatorial Optimization",

abstract = "We consider the approximate solution of discrete optimization problems using procedures that are capable of magnifying the effectiveness of any given heuristic algorithm through sequential application. In particular, we embed the problem within a dynamic programming framework, and we introduce several types of rollout algorithms, which are related to notions of policy iteration. We provide conditions guaranteeing that the rollout algorithm improves the performance of the original heuristic algorithm. The method is illustrated in the context of a machine maintenance and repair problem.",

author = "Bertsekas, {Dimitri P.} and Tsitsiklis, {John N.} and Cynara Wu",

note = "Funding Information: ⁄Research supported by NSF under Grant DMI-9625489.",

year = "1997",

doi = "10.1023/A:1009635226865",

language = "English (US)",

volume = "3",

pages = "245--262",

journal = "Journal of Heuristics",

issn = "1381-1231",

publisher = "Springer Netherlands",

number = "3",

}

TY - JOUR

T1 - Rollout Algorithms for Combinatorial Optimization

AU - Bertsekas, Dimitri P.

AU - Tsitsiklis, John N.

AU - Wu, Cynara

N1 - Funding Information: ⁄Research supported by NSF under Grant DMI-9625489.

PY - 1997

Y1 - 1997

N2 - We consider the approximate solution of discrete optimization problems using procedures that are capable of magnifying the effectiveness of any given heuristic algorithm through sequential application. In particular, we embed the problem within a dynamic programming framework, and we introduce several types of rollout algorithms, which are related to notions of policy iteration. We provide conditions guaranteeing that the rollout algorithm improves the performance of the original heuristic algorithm. The method is illustrated in the context of a machine maintenance and repair problem.

AB - We consider the approximate solution of discrete optimization problems using procedures that are capable of magnifying the effectiveness of any given heuristic algorithm through sequential application. In particular, we embed the problem within a dynamic programming framework, and we introduce several types of rollout algorithms, which are related to notions of policy iteration. We provide conditions guaranteeing that the rollout algorithm improves the performance of the original heuristic algorithm. The method is illustrated in the context of a machine maintenance and repair problem.

UR - http://www.scopus.com/inward/record.url?scp=0031272681&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031272681&partnerID=8YFLogxK

U2 - 10.1023/A:1009635226865

DO - 10.1023/A:1009635226865

M3 - Article

AN - SCOPUS:0031272681

VL - 3

SP - 245

EP - 262

JO - Journal of Heuristics

JF - Journal of Heuristics

SN - 1381-1231

IS - 3

ER -