Search results for "Uncertainty quantification"

showing 9 items of 29 documents

Leveraging Uncertainty Estimates to Improve Segmentation Performance in Cardiac MR

2021

In medical image segmentation, several studies have used Bayesian neural networks to segment and quantify the uncertainty of the images. These studies show that there might be an increased epistemic uncertainty in areas where there are semantically and visually challenging pixels. The uncertain areas of the image can be of a great interest as they can possibly indicate the regions of incorrect segmentation. To leverage the uncertainty information, we propose a segmentation model that incorporates the uncertainty into its learning process. Firstly, we generate the uncertainty estimate (sample variance) using Monte-Carlo dropout during training. Then we incorporate it into the loss function t…

PixelCalibration (statistics)business.industryComputer scienceComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONPattern recognitionImage segmentationLeverage (statistics)SegmentationSample varianceArtificial intelligenceUncertainty quantificationbusinessDropout (neural networks)
researchProduct

Algorithmic differentiation for cloud schemes (IFS Cy43r3) using CoDiPack (v1.8.1)

2019

Abstract. Numerical models in atmospheric sciences not only need to approximate the flow equations on a suitable computational grid, they also need to include subgrid effects of many non-resolved physical processes. Among others, the formation and evolution of cloud particles is an example of such subgrid processes. Moreover, to date there is no universal mathematical description of a cloud, hence many cloud schemes have been proposed and these schemes typically contain several uncertain parameters. In this study, we propose the use of algorithmic differentiation (AD) as a method to identify parameters within the cloud scheme, to which the output of the cloud scheme is most sensitive. We il…

Scheme (programming language)Mathematical optimization010504 meteorology & atmospheric sciencesComputer scienceAutomatic differentiationbusiness.industrylcsh:QE1-996.5Cloud computing010103 numerical & computational mathematicsGeneral MedicineLimitingNumerical modelsGrid01 natural scienceslcsh:GeologyFlow (mathematics)0101 mathematicsUncertainty quantificationbusinesscomputer0105 earth and related environmental sciencescomputer.programming_languageGeoscientific Model Development
researchProduct

Algorithmic Differentiation for Cloud Schemes

2019

<p>Numerical models in atmospheric sciences do not only need to approximate the flow equations on a suitable computational grid, they also need to include subgrid effects of many non-resolved physical processes. Among others, the formation and evolution of cloud particles is an example of such subgrid processes. Moreover, to date there is no universal mathematical description of a cloud, hence many cloud schemes were proposed and these schemes typically contain several uncertain parameters. In this study, we propose the use of algorithmic differentiation (AD) as a method to identify parameters within the cloud scheme, to which the output of the cloud scheme is most sensitive.…

Scheme (programming language)Mathematical optimizationAutomatic differentiationbusiness.industryComputer scienceCloud computingLimitingNumerical modelsGridFlow (mathematics)Uncertainty quantificationbusinesscomputercomputer.programming_language
researchProduct

A probabilistic compressive sensing framework with applications to ultrasound signal processing

2019

Abstract The field of Compressive Sensing (CS) has provided algorithms to reconstruct signals from a much lower number of measurements than specified by the Nyquist-Shannon theorem. There are two fundamental concepts underpinning the field of CS. The first is the use of random transformations to project high-dimensional measurements onto a much lower-dimensional domain. The second is the use of sparse regression to reconstruct the original signal. This assumes that a sparse representation exists for this signal in some known domain, manifested by a dictionary. The original formulation for CS specifies the use of an l 1 penalised regression method, the Lasso. Whilst this has worked well in l…

Signal processing0209 industrial biotechnologyBayesian methodsComputer scienceTKAerospace Engineering02 engineering and technologycomputer.software_genre01 natural sciencesRelevance vector machineNDTSettore ING-IND/14 - Progettazione Meccanica E Costruzione Di Macchine020901 industrial engineering & automationLasso (statistics)0103 physical sciencesUltrasoundUncertainty quantification010301 acousticsSparse representationCivil and Structural EngineeringSignal processingSignal reconstructionMechanical EngineeringProbabilistic logicSparse approximationCompressive sensingComputer Science ApplicationsCompressed sensingControl and Systems EngineeringRelevance Vector MachineData miningcomputer
researchProduct

Error Estimates of Theoretical Models: a Guide

2014

This guide offers suggestions/insights on uncertainty quantification of nuclear structure models. We discuss a simple approach to statistical error estimates, strategies to assess systematic errors, and show how to uncover inter-dependencies by correlation analysis. The basic concepts are illustrated through simple examples. By providing theoretical error bars on predicted quantities and using statistical methods to study correlations between observables, theory can significantly enhance the feedback between experiment and nuclear modeling.

Systematic errorPhysicsNuclear and High Energy PhysicsNuclear TheoryNuclear structureTheoretical modelsFOS: Physical sciencesObservableNuclear Theory (nucl-th)Simple (abstract algebra)Error barCorrelation analysisStatistical physicsUncertainty quantification
researchProduct

Herramientas matemáticas multi-escala para el tratamiento de señales

2019

En aquesta tesi doctoral, presentada com a compendi d'articles, es desenvolupen diverses eines matemàtiques per a processar senyals procedents de problemes reals. Aquestes tècniques estan basades en la teoria de la multi-resolució de Harten i la teoria d'esquemes de subdivisió. En aquest compendi es recullen treballs de disseny de nous esquemes de subdivisió no-lineals i aplicacions en Química Analítica, en optimització i en el disseny de velers de competició, entre d'altres. In this doctoral thesis, presented as a compendium of articles, various mathematical tools are developed to process signals from real problems. These techniques are based on Harten's multi-resolution framework and subd…

Tratamiento de señalesOptimización no-linealQuímica AnalíticaUncertainty QuantificationEsquemas de subdivisiónAnálisis de multirresoluciónMultirresolución de HartenCromatografíaAproximación numérica
researchProduct

High-accuracy approximation of piecewise smooth functions using the Truncation and Encode approach

2017

Abstract In the present work, we analyze a technique designed by Geraci et al. in [1,11] named the Truncate and Encode (TE) strategy. It was presented as a non-intrusive method for steady and non-steady Partial Differential Equations (PDEs) in Uncertainty Quantification (UQ), and as a weakly intrusive method in the unsteady case. We analyze the TE algorithm applied to the approximation of functions, and in particular its performance for piecewise smooth functions. We carry out some numerical experiments, comparing the performance of the algorithm when using different linear and non-linear interpolation techniques and provide some recommendations that we find useful in order to achieve a hig…

Truncation errorPartial differential equationGeneral Computer ScienceTruncationApplied MathematicsMathematical analysisOrder (ring theory)010103 numerical & computational mathematicsENCODE01 natural sciences010101 applied mathematicsModeling and SimulationPiecewiseApplied mathematics0101 mathematicsUncertainty quantificationEngineering (miscellaneous)InterpolationApplied Mathematics and Nonlinear Sciences
researchProduct

Automated uncertainty quantification analysis using a system model and data

2015

International audience; Understanding the sources of, and quantifying the magnitude of, uncertainty can improve decision-making and, thereby, make manufacturing systems more efficient. Achieving this goal requires knowledge in two separate domains: data science and manufacturing. In this paper, we focus on quantifying uncertainty, usually called uncertainty quantification (UQ). More specifically, we propose a methodology to perform UQ automatically using Bayesian networks (BN) constructed from three types of sources: a descriptive system model, physics-based mathematical models, and data. The system model is a high-level model describing the system and its parameters; we develop this model …

[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI]generic modeling environment[SPI] Engineering Sciences [physics]Computer scienceuncertainty quantificationMachine learningcomputer.software_genre01 natural sciencesData modelingSystem model[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]010104 statistics & probability03 medical and health sciences[SPI]Engineering Sciences [physics][ SPI ] Engineering Sciences [physics]Sensitivity analysis0101 mathematicsUncertainty quantification[ INFO.INFO-AI ] Computer Science [cs]/Artificial Intelligence [cs.AI]030304 developmental biologyautomation0303 health sciencesMathematical modelbusiness.industryConditional probabilityBayesian networkmeta-modelMetamodelingBayesian networkProbability distributionData miningArtificial intelligencebusinesscomputer
researchProduct

Uncertainty in urban stormwater quality modelling: The effect of acceptability threshold in the GLUE methodology

2007

Uncertainty analysis in integrated urban drainage modelling is of growing importance in the field of water quality. However, only few studies deal with uncertainty quantification in urban drainage modelling; furthermore, the few existing studies mainly focus on quantitative sewer flow modelling rather than uncertainty in water quality aspects. In this context, the generalised likelihood uncertainty estimation (GLUE) methodology was applied for the evaluation of the uncertainty of an integrated urban drainage model and some of its subjective hypotheses have been explored. More specifically, the influence of the subjective choice of the acceptability threshold has been detected in order to ga…

geographyEnvironmental Engineeringgeography.geographical_feature_categoryComputer scienceEcological ModelingStormwaterUncertaintyEnvironmental engineeringContext (language use)Models TheoreticalUrban areaPollutionWater SupplyEconometricsWater qualityDrainageUncertainty quantificationGLUEWaste Management and DisposalUncertainty analysisWater Science and TechnologyCivil and Structural Engineering
researchProduct