4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering最新文献_第3页

EFFECTIVENESS OF THE PROBABILITY DENSITY EVOLUTION METHOD FOR DYNAMIC AND RELIABILITY ANALYSES OF MASONRY STRUCTURES 概率密度演化法在砌体结构动力和可靠度分析中的有效性

4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering Pub Date : 1900-01-01 DOI: 10.7712/120221.8040.19025

M. Lucchesi, B. Pintucchi, N. Zani

引用次数: 1

BOUNDS OF RELIABILITY FUNCTION FOR STRUCTURAL SYSTEMS SUBJECTED TO A SET OF RECORDED ACCELEROGRAMS 结构系统在一组记录加速度作用下的可靠度函数边界

4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering Pub Date : 1900-01-01 DOI: 10.7712/120221.8020.18940

F. Genovese, G. Muscolino, A. Sofi

引用次数: 0

SENSOR FAULT IDENTIFICATION FOR ROBUST STRUCTURAL HEALTH MONITORING 用于鲁棒结构健康监测的传感器故障识别

4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering Pub Date : 1900-01-01 DOI: 10.7712/120221.8029.19011

Andreea-Maria Oncescu, A. Cicirello

引用次数: 0

ADAPTIVE SEQUENTIAL SAMPLING FOR POLYNOMIAL CHAOS EXPANSION 多项式混沌展开的自适应顺序采样

4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering Pub Date : 1900-01-01 DOI: 10.7712/120221.8038.18955

L. Novák, M. Vořechovský, Václav Sadílek

引用次数: 0

UNCERTAINTY QUANTIFICATION FOR DEEP LEARNING REGRESSION MODELS IN THE LOW DATA LIMIT 低数据极限下深度学习回归模型的不确定性量化

4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering Pub Date : 1900-01-01 DOI: 10.7712/120221.8045.19145

Cristina Garcia-Cardona, Yen-Ting Lin, T. Bhattacharya

{"title":"UNCERTAINTY QUANTIFICATION FOR DEEP LEARNING REGRESSION MODELS IN THE LOW DATA LIMIT","authors":"Cristina Garcia-Cardona, Yen-Ting Lin, T. Bhattacharya","doi":"10.7712/120221.8045.19145","DOIUrl":"https://doi.org/10.7712/120221.8045.19145","url":null,"abstract":". Deep learning models have contributed to a broad range of applications, but require large amounts of data to learn the desired input-output mapping. Despite the success in developing prediction engines that have high accuracy, much less attention has been given to assessing the error associated with individual predictions. In this work, we study machine-learning models of uncertainty quantiﬁcation for regression, i.e., methods that are almost purely data driven and use deep learning itself to quantify the conﬁdence in its predictions. We use two approaches, namely the heteroscedastic and quantile formulations, and their extensions to problems with multidimensional output. We focus on the low data limit, where the data sets available are on the order of hundred, not thousands, samples. Through numerical experiments we demonstrate that both heteroscedastic and quantile formulations are robust and good at uncertainty estimation even in this low data limit. We note that the quantile formulation seems to have better performance and is more stable than the heteroscedastic case. Overall, our studies pave the way towards practical design of deep learning models that provide actionable predictions with quantiﬁed uncertainty using accessible volumes of data.","PeriodicalId":444608,"journal":{"name":"4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127146437","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

CALIBRATION OF MATERIAL MODEL PARAMETERS USING MIXED-EFFECTS MODEL 使用混合效应模型校正材料模型参数

4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering Pub Date : 1900-01-01 DOI: 10.7712/120221.8037.18933

Clément Laboulfie, Matthieu Balesdent, Loïc Brevault, S. Da Veiga, F. Irisarri, Rodolphe Le Riche, J. Maire

{"title":"CALIBRATION OF MATERIAL MODEL PARAMETERS USING MIXED-EFFECTS MODEL","authors":"Clément Laboulfie, Matthieu Balesdent, Loïc Brevault, S. Da Veiga, F. Irisarri, Rodolphe Le Riche, J. Maire","doi":"10.7712/120221.8037.18933","DOIUrl":"https://doi.org/10.7712/120221.8037.18933","url":null,"abstract":"The quantification of model parameter uncertainty is a long-standing issue in model calibration. Classical techniques provide methods to handle some type of uncertainties (e.g. experimental noise or model bias). However, usual calibration techniques are not designed to take into account the variability between the different individuals. This is not a problem if the individual variability is negligible but it is an important issue if the individual variability is signifcant. The mixed-effects models provide a statistical framework to calibrate the parameters of a model taking into account the individual variability. The objective of this paper is to introduce the mixed-effects in material science. The ONERA Damage model (ODM) is considered, first with synthetic data, then with thirteen experimental strain-stress curves of a ceramic matrix composite material. The robustness of the mixed-effects approach regarding the variability and the number of specimen is investigated. Model choices such as the correlation between ODM parameters and other settings are discussed. The ability of mixed-effects models to characterize the material variability and to provide accurate estimates of the parameters associated to each specimen is illustrated.","PeriodicalId":444608,"journal":{"name":"4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133450892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

OTBENCHMARK: AN OPEN SOURCE PYTHON PACKAGE FOR BENCHMARKING AND VALIDATING UNCERTAINTY QUANTIFICATION ALGORITHMS Otbenchmark:用于基准测试和验证不确定性量化算法的开源python包

4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering Pub Date : 1900-01-01 DOI: 10.7712/120221.8034.19093

E. Fekhari, M. Baudin, V. Chabridon, Youssef Jebroun

{"title":"OTBENCHMARK: AN OPEN SOURCE PYTHON PACKAGE FOR BENCHMARKING AND VALIDATING UNCERTAINTY QUANTIFICATION ALGORITHMS","authors":"E. Fekhari, M. Baudin, V. Chabridon, Youssef Jebroun","doi":"10.7712/120221.8034.19093","DOIUrl":"https://doi.org/10.7712/120221.8034.19093","url":null,"abstract":". Over the past decade, industrial companies and academic institutions pooled their efforts and knowledge to propose a generic uncertainty management methodology for computer simulation. This framework led to the collaborative development of an open source software dedicated to the treatment of uncertainties, called “OpenTURNS” (Open source Treatment of Uncertainty, Risk’N Statistics). This paper aims at presenting a new Python package, called “ otbenchmark ”, offering tools to evaluate the performance of a large panel of uncertainty quantiﬁcation algorithms. It provides benchmark classes containing problems with their reference values. Two categories of benchmark classes are currently available: reliability estimation problems ( i.e., estimating failure probabilities) and sensitivity analysis problems ( i.e., estimating sensitivity indices such as the Sobol’ indices). This package can either be used for validating a new algorithm or automatically comparing various algorithms on a set of problems. Additionally, the package provides several convergence and accuracy metrics to compare the performance of each algorithm. To face high-dimensional problems, otbenchmark offers graphical tools to draw multidimensional events, functions and distributions based on cross-cuts visualizations. Finally, to ensure otbenchmark ’s accuracy, a test-driven software development method has been adopted (using, among others, Git for collaborative development, unit tests and continuous integration). Ultimately, otbenchmark is an industrial platform gath-ering problems with reference values of their solutions and various tools to achieve a robust comparison of uncertainty management algorithms.","PeriodicalId":444608,"journal":{"name":"4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117048482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

STRUCTURAL RELIABILITY ESTIMATION OF STEEL MAST EXHIBITING RANDOM MECHANICAL AND ENVIRONMENTAL PARAMETERS 具有随机力学和环境参数的钢桅杆结构可靠度估计

4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering Pub Date : 1900-01-01 DOI: 10.7712/120221.8021.18961

R. Bredow, M. Kaminski

引用次数: 0