Estimation of an improved surrogate model in uncertainty quantification by neural networks

被引：0

作者：

Benedict Götz

Sebastian Kersting

Michael Kohler

机构：

[1] Technische Universität Darmstadt,Fachgebiet Systemzuverlässigkeit, Adaptronik und Maschinenakustik SAM

[2] Technische Universität Darmstadt,Fachbereich Mathematik

来源：

Annals of the Institute of Statistical Mathematics | 2021年 / 73卷

关键词：

Curse of dimensionality; Density estimation; Imperfect models; error; Neural networks; Surrogate models; Uncertainty quantification;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

Quantification of uncertainty of a technical system is often based on a surrogate model of a corresponding simulation model. In any application, the simulation model will not describe the reality perfectly, and consequently the surrogate model will be imperfect. In this article, we combine observed data from the technical system with simulated data from the imperfect simulation model in order to estimate an improved surrogate model consisting of multilayer feedforward neural networks, and we show that under suitable assumptions, this estimate is able to circumvent the curse of dimensionality. Based on this improved surrogate model, we show a rate of the convergence result for density estimates. The finite sample size performance of the estimates is illustrated by applying them to simulated data. The practical usefulness of the newly proposed estimates is demonstrated by using them to predict the uncertainty of a lateral vibration attenuation system with piezo-elastic supports.

引用

页码：249 / 281

页数：32

共 50 条

[31] Neural orientation distribution fields for estimation and uncertainty quantification in diffusion MRI [J].

Consagra, William ;

Ning, Lipeng ;

Rathi, Yogesh .

MEDICAL IMAGE ANALYSIS, 2024, 93

[32] Global Surrogate Modeling by Neural Network-Based Model Uncertainty [J].

Leifsson, Leifur ;

Nagawkar, Jethro ;

Barnet, Laurel ;

Bryden, Kenneth ;

Koziel, Slawomir ;

Pietrenko-Dabrowska, Anna .

COMPUTATIONAL SCIENCE - ICCS 2022, PT III, 2022, 13352 :425-434

[33] Epistemic Uncertainty Quantification in State-Space LPV Model Identification Using Bayesian Neural Networks [J].

Bao, Yajie ;

Velni, Javad Mohammadpour ;

Shahbakhti, Mahdi .

IEEE CONTROL SYSTEMS LETTERS, 2021, 5 (02) :719-724

[34] Efficient uncertainty quantification for dynamic subsurface flow with surrogate by Theory-guided Neural Network [J].

Wang, Nanzhe ;

Chang, Haibin ;

Zhang, Dongxiao .

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 2021, 373

[35] Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification [J].

Tripathy, Rohit K. ;

Bilionis, Ilias .

JOURNAL OF COMPUTATIONAL PHYSICS, 2018, 375 :565-588

[36] Uncertainty Quantification in Inverse Scattering Problems With Bayesian Convolutional Neural Networks [J].

Wei, Zhun ;

Chen, Xudong .

IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, 2021, 69 (06) :3409-3418

[37] HatchEnsemble: an efficient and practical uncertainty quantification method for deep neural networks [J].

Xia, Yufeng ;

Zhang, Jun ;

Jiang, Tingsong ;

Gong, Zhiqiang ;

Yao, Wen ;

Feng, Ling .

COMPLEX & INTELLIGENT SYSTEMS, 2021, 7 (06) :2855-2869

[38] Uncertainty quantification for deep neural networks: An empirical comparison and usage guidelines [J].

Weiss, Michael ;

Tonella, Paolo .

SOFTWARE TESTING VERIFICATION & RELIABILITY, 2023, 33 (06)

[39] Density regression and uncertainty quantification with Bayesian deep noise neural networks [J].

Zhang, Daiwei ;

Liu, Tianci ;

Kang, Jian .

STAT, 2023, 12 (01)

[40] HatchEnsemble: an efficient and practical uncertainty quantification method for deep neural networks [J].

Yufeng Xia ;

Jun Zhang ;

Tingsong Jiang ;

Zhiqiang Gong ;

Wen Yao ;

Ling Feng .

Complex & Intelligent Systems, 2021, 7 :2855-2869

← 1 2 3 4 5 →