Approximate implementations of pure random search in the presence of noise

David L.J. Alexander; David W. Bulger; James M. Calvin; H. Edwin Romeijn; Ryan L. Sherriff

doi:10.1007/s10898-004-9970-4

Approximate implementations of pure random search in the presence of noise

David L.J. Alexander, David W. Bulger, James M. Calvin, H. Edwin Romeijn, Ryan L. Sherriff

Computer Science

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

6 Scopus citations

Abstract

We discuss the noisy optimisation problem, in which function evaluations are subject to random noise. Adaptation of pure random search to noisy optimisation by repeated sampling is considered. We introduce and exploit an improving bias condition on noise-affected pure random search algorithms. Two such algorithms are considered; we show that one requires infinite expected work to proceed, while the other is practical.

Original language	English (US)
Pages (from-to)	601-612
Number of pages	12
Journal	Journal of Global Optimization
Volume	31
Issue number	4
DOIs	https://doi.org/10.1007/s10898-004-9970-4
State	Published - Apr 2005

All Science Journal Classification (ASJC) codes

Control and Optimization
Applied Mathematics
Business, Management and Accounting (miscellaneous)
Computer Science Applications
Management Science and Operations Research

Keywords

Global optimisation
Noisy objective function
Pure random search
Sequential analysis

Access to Document

10.1007/s10898-004-9970-4

Cite this

@article{71e262d45bf84450aa10b85feae9dae1,

title = "Approximate implementations of pure random search in the presence of noise",

abstract = "We discuss the noisy optimisation problem, in which function evaluations are subject to random noise. Adaptation of pure random search to noisy optimisation by repeated sampling is considered. We introduce and exploit an improving bias condition on noise-affected pure random search algorithms. Two such algorithms are considered; we show that one requires infinite expected work to proceed, while the other is practical.",

keywords = "Global optimisation, Noisy objective function, Pure random search, Sequential analysis",

author = "Alexander, {David L.J.} and Bulger, {David W.} and Calvin, {James M.} and Romeijn, {H. Edwin} and Sherriff, {Ryan L.}",

note = "Funding Information: ?Supported by a Bright Future Scholarship administered by the Foundation for Research, Science and Technology, New Zealand. Funding Information: This paper is the result of work conducted at the SGO 2001 Workshop in Hanmer, New Zealand, June 2001, funded in part by the Marsden Fund of the Royal Society of New Zealand.",

year = "2005",

month = apr,

doi = "10.1007/s10898-004-9970-4",

language = "English (US)",

volume = "31",

pages = "601--612",

journal = "Journal of Global Optimization",

issn = "0925-5001",

publisher = "Springer Netherlands",

number = "4",

}

TY - JOUR

T1 - Approximate implementations of pure random search in the presence of noise

AU - Alexander, David L.J.

AU - Bulger, David W.

AU - Calvin, James M.

AU - Romeijn, H. Edwin

AU - Sherriff, Ryan L.

N1 - Funding Information: ?Supported by a Bright Future Scholarship administered by the Foundation for Research, Science and Technology, New Zealand. Funding Information: This paper is the result of work conducted at the SGO 2001 Workshop in Hanmer, New Zealand, June 2001, funded in part by the Marsden Fund of the Royal Society of New Zealand.

PY - 2005/4

Y1 - 2005/4

N2 - We discuss the noisy optimisation problem, in which function evaluations are subject to random noise. Adaptation of pure random search to noisy optimisation by repeated sampling is considered. We introduce and exploit an improving bias condition on noise-affected pure random search algorithms. Two such algorithms are considered; we show that one requires infinite expected work to proceed, while the other is practical.

AB - We discuss the noisy optimisation problem, in which function evaluations are subject to random noise. Adaptation of pure random search to noisy optimisation by repeated sampling is considered. We introduce and exploit an improving bias condition on noise-affected pure random search algorithms. Two such algorithms are considered; we show that one requires infinite expected work to proceed, while the other is practical.

KW - Global optimisation

KW - Noisy objective function

KW - Pure random search

KW - Sequential analysis

UR - http://www.scopus.com/inward/record.url?scp=22044433804&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=22044433804&partnerID=8YFLogxK

U2 - 10.1007/s10898-004-9970-4

DO - 10.1007/s10898-004-9970-4

M3 - Article

AN - SCOPUS:22044433804

SN - 0925-5001

VL - 31

SP - 601

EP - 612

JO - Journal of Global Optimization

JF - Journal of Global Optimization

IS - 4

ER -

Approximate implementations of pure random search in the presence of noise

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this