Surrogate modeling for sensitivity analysis of models with high-dimensional outputs

Min Li; Gaofeng Jia; Ruo Qian Wang

doi:10.22725/ICASP13.280

Surrogate modeling for sensitivity analysis of models with high-dimensional outputs

Min Li, Gaofeng Jia, Ruo Qian Wang

Science and Engineering Office

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Sensitivity analysis provides important information on how the input uncertainty impacts the system output uncertainty. Typically, sensitivity analysis entails large number of system evaluations. For expensive system models with high-dimensional outputs, direct adoption of such models for sensitivity analysis poses significant computational challenges. To address these challenges, an efficient dimension reduction and surrogate based approach is proposed for efficient sensitivity analysis of expensive system models with high-dimensional outputs. As an example, the proposed approach is applied to investigate the sensitivity of peak water level over large coastal regions in San Francisco Bay with respect to the construction of levees at different counties under projected sea level rise.

Original language	English
Title of host publication	ICASP13 Proceedings
Place of Publication	Seoul
Publisher	Seoul National University
Pages	1-8
Number of pages	8
DOIs	https://doi.org/10.22725/ICASP13.280
Publication status	Published - 26 May 2019
Event	13th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP 2019 - Seoul, Korea, Republic of Duration: 26 May 2019 → 30 May 2019

Conference

Conference	13th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP 2019
Country/Territory	Korea, Republic of
City	Seoul
Period	26/05/19 → 30/05/19

ASJC Scopus subject areas

Civil and Structural Engineering
Statistics and Probability

Access to Document

10.22725/ICASP13.280

Cite this

@inproceedings{f0278b58398047039a2e56b68d407321,

title = "Surrogate modeling for sensitivity analysis of models with high-dimensional outputs",

abstract = "Sensitivity analysis provides important information on how the input uncertainty impacts the system output uncertainty. Typically, sensitivity analysis entails large number of system evaluations. For expensive system models with high-dimensional outputs, direct adoption of such models for sensitivity analysis poses significant computational challenges. To address these challenges, an efficient dimension reduction and surrogate based approach is proposed for efficient sensitivity analysis of expensive system models with high-dimensional outputs. As an example, the proposed approach is applied to investigate the sensitivity of peak water level over large coastal regions in San Francisco Bay with respect to the construction of levees at different counties under projected sea level rise.",

author = "Min Li and Gaofeng Jia and Wang, {Ruo Qian}",

year = "2019",

month = may,

day = "26",

doi = "10.22725/ICASP13.280",

language = "English",

pages = "1--8",

booktitle = "ICASP13 Proceedings",

publisher = "Seoul National University",

address = "Korea, Republic of",

note = "13th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP 2019 ; Conference date: 26-05-2019 Through 30-05-2019",

}

Li, M, Jia, G & Wang, RQ 2019, Surrogate modeling for sensitivity analysis of models with high-dimensional outputs. in ICASP13 Proceedings., 280, Seoul National University, Seoul, pp. 1-8, 13th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP 2019, Seoul, Korea, Republic of, 26/05/19. https://doi.org/10.22725/ICASP13.280

TY - GEN

T1 - Surrogate modeling for sensitivity analysis of models with high-dimensional outputs

AU - Li, Min

AU - Jia, Gaofeng

AU - Wang, Ruo Qian

PY - 2019/5/26

Y1 - 2019/5/26

N2 - Sensitivity analysis provides important information on how the input uncertainty impacts the system output uncertainty. Typically, sensitivity analysis entails large number of system evaluations. For expensive system models with high-dimensional outputs, direct adoption of such models for sensitivity analysis poses significant computational challenges. To address these challenges, an efficient dimension reduction and surrogate based approach is proposed for efficient sensitivity analysis of expensive system models with high-dimensional outputs. As an example, the proposed approach is applied to investigate the sensitivity of peak water level over large coastal regions in San Francisco Bay with respect to the construction of levees at different counties under projected sea level rise.

AB - Sensitivity analysis provides important information on how the input uncertainty impacts the system output uncertainty. Typically, sensitivity analysis entails large number of system evaluations. For expensive system models with high-dimensional outputs, direct adoption of such models for sensitivity analysis poses significant computational challenges. To address these challenges, an efficient dimension reduction and surrogate based approach is proposed for efficient sensitivity analysis of expensive system models with high-dimensional outputs. As an example, the proposed approach is applied to investigate the sensitivity of peak water level over large coastal regions in San Francisco Bay with respect to the construction of levees at different counties under projected sea level rise.

UR - http://www.scopus.com/inward/record.url?scp=85071022062&partnerID=8YFLogxK

U2 - 10.22725/ICASP13.280

DO - 10.22725/ICASP13.280

M3 - Conference contribution

SP - 1

EP - 8

BT - ICASP13 Proceedings

PB - Seoul National University

CY - Seoul

T2 - 13th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP 2019

Y2 - 26 May 2019 through 30 May 2019

ER -

Surrogate modeling for sensitivity analysis of models with high-dimensional outputs

Abstract

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this