Fast predictive multi-fidelity prediction with models of quantized fidelity levels

Mani Razi, Robert Michael Ii Kirby, Akil Narayan

Research output: Contribution to journalArticle

Abstract

In this paper, we introduce a novel approach for the construction of multi-fidelity surrogate models with “discrete” fidelity levels. The notion of a discrete level of fidelity is in contrast to a mathematical model, for which the notion of refinement towards a high-fidelity model is relevant to sending a discretization parameter toward zero in a continuous way. Our notion of discrete fidelity levels encompasses cases for which there is no notion of convergence in terms of a fidelity parameter that can be sent to zero or infinity. The particular choice of how levels of fidelity are defined in this framework paves the way for using models that may have no apparent physical or mathematical relationship to the target high-fidelity model. However, our approach requires that models can produce results with a common set of parameters in the target model. Hence, fidelity level in this work is not directly representative of the degree of similarity of a low-fidelity model to a target high-fidelity model. In particular, we show that our approach is applicable to competitive ecological systems with different numbers of species, discrete-state Markov chains with a different number of states, polymer networks with a different number of connections, and nano-particle plasmonic arrays with a different number of scatterers. The results of this study demonstrate that our procedure boasts computational efficiency and accuracy for a wide variety of models and engineering systems.

LanguageEnglish (US)
Pages992-1008
Number of pages17
JournalJournal of Computational Physics
Volume376
DOIs
StatePublished - Jan 1 2019

Fingerprint

Parameter estimation
predictions
Computational efficiency
Systems engineering
Markov processes
Markov chains
ecosystems
systems engineering
infinity
Mathematical models
mathematical models
Polymers

Keywords

  • Discrete systems
  • Model-order reduction
  • Multi-fidelity models
  • Parameter estimation
  • Surrogate modeling

ASJC Scopus subject areas

  • Physics and Astronomy (miscellaneous)
  • Computer Science Applications

Cite this

Fast predictive multi-fidelity prediction with models of quantized fidelity levels. / Razi, Mani; Kirby, Robert Michael Ii; Narayan, Akil.

In: Journal of Computational Physics, Vol. 376, 01.01.2019, p. 992-1008.

Research output: Contribution to journalArticle

@article{93777945d4d64e3c8187c6f0245278b1,
title = "Fast predictive multi-fidelity prediction with models of quantized fidelity levels",
abstract = "In this paper, we introduce a novel approach for the construction of multi-fidelity surrogate models with “discrete” fidelity levels. The notion of a discrete level of fidelity is in contrast to a mathematical model, for which the notion of refinement towards a high-fidelity model is relevant to sending a discretization parameter toward zero in a continuous way. Our notion of discrete fidelity levels encompasses cases for which there is no notion of convergence in terms of a fidelity parameter that can be sent to zero or infinity. The particular choice of how levels of fidelity are defined in this framework paves the way for using models that may have no apparent physical or mathematical relationship to the target high-fidelity model. However, our approach requires that models can produce results with a common set of parameters in the target model. Hence, fidelity level in this work is not directly representative of the degree of similarity of a low-fidelity model to a target high-fidelity model. In particular, we show that our approach is applicable to competitive ecological systems with different numbers of species, discrete-state Markov chains with a different number of states, polymer networks with a different number of connections, and nano-particle plasmonic arrays with a different number of scatterers. The results of this study demonstrate that our procedure boasts computational efficiency and accuracy for a wide variety of models and engineering systems.",
keywords = "Discrete systems, Model-order reduction, Multi-fidelity models, Parameter estimation, Surrogate modeling",
author = "Mani Razi and Kirby, {Robert Michael Ii} and Akil Narayan",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.jcp.2018.10.025",
language = "English (US)",
volume = "376",
pages = "992--1008",
journal = "Journal of Computational Physics",
issn = "0021-9991",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Fast predictive multi-fidelity prediction with models of quantized fidelity levels

AU - Razi, Mani

AU - Kirby, Robert Michael Ii

AU - Narayan, Akil

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In this paper, we introduce a novel approach for the construction of multi-fidelity surrogate models with “discrete” fidelity levels. The notion of a discrete level of fidelity is in contrast to a mathematical model, for which the notion of refinement towards a high-fidelity model is relevant to sending a discretization parameter toward zero in a continuous way. Our notion of discrete fidelity levels encompasses cases for which there is no notion of convergence in terms of a fidelity parameter that can be sent to zero or infinity. The particular choice of how levels of fidelity are defined in this framework paves the way for using models that may have no apparent physical or mathematical relationship to the target high-fidelity model. However, our approach requires that models can produce results with a common set of parameters in the target model. Hence, fidelity level in this work is not directly representative of the degree of similarity of a low-fidelity model to a target high-fidelity model. In particular, we show that our approach is applicable to competitive ecological systems with different numbers of species, discrete-state Markov chains with a different number of states, polymer networks with a different number of connections, and nano-particle plasmonic arrays with a different number of scatterers. The results of this study demonstrate that our procedure boasts computational efficiency and accuracy for a wide variety of models and engineering systems.

AB - In this paper, we introduce a novel approach for the construction of multi-fidelity surrogate models with “discrete” fidelity levels. The notion of a discrete level of fidelity is in contrast to a mathematical model, for which the notion of refinement towards a high-fidelity model is relevant to sending a discretization parameter toward zero in a continuous way. Our notion of discrete fidelity levels encompasses cases for which there is no notion of convergence in terms of a fidelity parameter that can be sent to zero or infinity. The particular choice of how levels of fidelity are defined in this framework paves the way for using models that may have no apparent physical or mathematical relationship to the target high-fidelity model. However, our approach requires that models can produce results with a common set of parameters in the target model. Hence, fidelity level in this work is not directly representative of the degree of similarity of a low-fidelity model to a target high-fidelity model. In particular, we show that our approach is applicable to competitive ecological systems with different numbers of species, discrete-state Markov chains with a different number of states, polymer networks with a different number of connections, and nano-particle plasmonic arrays with a different number of scatterers. The results of this study demonstrate that our procedure boasts computational efficiency and accuracy for a wide variety of models and engineering systems.

KW - Discrete systems

KW - Model-order reduction

KW - Multi-fidelity models

KW - Parameter estimation

KW - Surrogate modeling

UR - http://www.scopus.com/inward/record.url?scp=85055545942&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055545942&partnerID=8YFLogxK

U2 - 10.1016/j.jcp.2018.10.025

DO - 10.1016/j.jcp.2018.10.025

M3 - Article

VL - 376

SP - 992

EP - 1008

JO - Journal of Computational Physics

T2 - Journal of Computational Physics

JF - Journal of Computational Physics

SN - 0021-9991

ER -