High-dimensional model representations generated from low order terms-lp-RS-HDMR

L. I. Genyuan, Maxim Artamonov, Herschel Rabitz, Sheng Wei Wang, Panagiotis Georgopoulos, Metin Demiralp

Research output: Contribution to journalArticle

36 Citations (Scopus)

Abstract

High-dimensional model representation (HDMR) is a general set of quantitative model assessment and analysis tools for improving the efficiency of deducing high dimensional input-output system behavior. RS-HDMR is a particular form of HDMR based on random sampling (RS) of the input variables. The component functions in an HDMR expansion are optimal choices tailored to the n-variate function f(x) being represented over the desired domain of the n-dimensional vector x. The high-order terms (usually larger than second order, or equivalently beyond cooperativity between pairs of variables) in the expansion are often negligible. When it is necessary to go beyond the first and the second order RS-HDMR, this article introduces a modified low-order term product (lp)-RS-HDMR method to approximately represent the high-order RS-HDMR component functions as products of low-order functions. Using this method the high-order truncated RS-HDMR expansions may be constructed without directly computing the original high-order terms. The mathematical foundations of lp-RS-HDMR are presented along with an illustration of its utility in an atmospheric chemical kinetics model.

Original languageEnglish (US)
Pages (from-to)647-656
Number of pages10
JournalJournal of Computational Chemistry
Volume24
Issue number5
DOIs
StatePublished - Apr 15 2003

Fingerprint

High Dimensional Model Representation
Random Sampling
Sampling
Term
Higher Order
Chemical Kinetics
Kinetic Model
n-dimensional
High-dimensional
Reaction kinetics

All Science Journal Classification (ASJC) codes

  • Computational Mathematics
  • Chemistry(all)

Keywords

  • Atmospheric modeling
  • HDMR
  • Monte Carlo integration
  • Random sampling

Cite this

Genyuan, L. I. ; Artamonov, Maxim ; Rabitz, Herschel ; Wang, Sheng Wei ; Georgopoulos, Panagiotis ; Demiralp, Metin. / High-dimensional model representations generated from low order terms-lp-RS-HDMR. In: Journal of Computational Chemistry. 2003 ; Vol. 24, No. 5. pp. 647-656.
@article{4a13674fb45149d4ad8b70bd5c3706c4,
title = "High-dimensional model representations generated from low order terms-lp-RS-HDMR",
abstract = "High-dimensional model representation (HDMR) is a general set of quantitative model assessment and analysis tools for improving the efficiency of deducing high dimensional input-output system behavior. RS-HDMR is a particular form of HDMR based on random sampling (RS) of the input variables. The component functions in an HDMR expansion are optimal choices tailored to the n-variate function f(x) being represented over the desired domain of the n-dimensional vector x. The high-order terms (usually larger than second order, or equivalently beyond cooperativity between pairs of variables) in the expansion are often negligible. When it is necessary to go beyond the first and the second order RS-HDMR, this article introduces a modified low-order term product (lp)-RS-HDMR method to approximately represent the high-order RS-HDMR component functions as products of low-order functions. Using this method the high-order truncated RS-HDMR expansions may be constructed without directly computing the original high-order terms. The mathematical foundations of lp-RS-HDMR are presented along with an illustration of its utility in an atmospheric chemical kinetics model.",
keywords = "Atmospheric modeling, HDMR, Monte Carlo integration, Random sampling",
author = "Genyuan, {L. I.} and Maxim Artamonov and Herschel Rabitz and Wang, {Sheng Wei} and Panagiotis Georgopoulos and Metin Demiralp",
year = "2003",
month = "4",
day = "15",
doi = "https://doi.org/10.1002/jcc.10232",
language = "English (US)",
volume = "24",
pages = "647--656",
journal = "Journal of Computational Chemistry",
issn = "0192-8651",
publisher = "John Wiley and Sons Inc.",
number = "5",

}

High-dimensional model representations generated from low order terms-lp-RS-HDMR. / Genyuan, L. I.; Artamonov, Maxim; Rabitz, Herschel; Wang, Sheng Wei; Georgopoulos, Panagiotis; Demiralp, Metin.

In: Journal of Computational Chemistry, Vol. 24, No. 5, 15.04.2003, p. 647-656.

Research output: Contribution to journalArticle

TY - JOUR

T1 - High-dimensional model representations generated from low order terms-lp-RS-HDMR

AU - Genyuan, L. I.

AU - Artamonov, Maxim

AU - Rabitz, Herschel

AU - Wang, Sheng Wei

AU - Georgopoulos, Panagiotis

AU - Demiralp, Metin

PY - 2003/4/15

Y1 - 2003/4/15

N2 - High-dimensional model representation (HDMR) is a general set of quantitative model assessment and analysis tools for improving the efficiency of deducing high dimensional input-output system behavior. RS-HDMR is a particular form of HDMR based on random sampling (RS) of the input variables. The component functions in an HDMR expansion are optimal choices tailored to the n-variate function f(x) being represented over the desired domain of the n-dimensional vector x. The high-order terms (usually larger than second order, or equivalently beyond cooperativity between pairs of variables) in the expansion are often negligible. When it is necessary to go beyond the first and the second order RS-HDMR, this article introduces a modified low-order term product (lp)-RS-HDMR method to approximately represent the high-order RS-HDMR component functions as products of low-order functions. Using this method the high-order truncated RS-HDMR expansions may be constructed without directly computing the original high-order terms. The mathematical foundations of lp-RS-HDMR are presented along with an illustration of its utility in an atmospheric chemical kinetics model.

AB - High-dimensional model representation (HDMR) is a general set of quantitative model assessment and analysis tools for improving the efficiency of deducing high dimensional input-output system behavior. RS-HDMR is a particular form of HDMR based on random sampling (RS) of the input variables. The component functions in an HDMR expansion are optimal choices tailored to the n-variate function f(x) being represented over the desired domain of the n-dimensional vector x. The high-order terms (usually larger than second order, or equivalently beyond cooperativity between pairs of variables) in the expansion are often negligible. When it is necessary to go beyond the first and the second order RS-HDMR, this article introduces a modified low-order term product (lp)-RS-HDMR method to approximately represent the high-order RS-HDMR component functions as products of low-order functions. Using this method the high-order truncated RS-HDMR expansions may be constructed without directly computing the original high-order terms. The mathematical foundations of lp-RS-HDMR are presented along with an illustration of its utility in an atmospheric chemical kinetics model.

KW - Atmospheric modeling

KW - HDMR

KW - Monte Carlo integration

KW - Random sampling

UR - http://www.scopus.com/inward/record.url?scp=0037447145&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037447145&partnerID=8YFLogxK

U2 - https://doi.org/10.1002/jcc.10232

DO - https://doi.org/10.1002/jcc.10232

M3 - Article

VL - 24

SP - 647

EP - 656

JO - Journal of Computational Chemistry

JF - Journal of Computational Chemistry

SN - 0192-8651

IS - 5

ER -