Representing model discrepancy in bound-to-bound data collaboration

DOI10.1137/19M1270185MaRDI QIDQ5858423zbMATH OpenWikidataFDO

Authors Wenyu Li, Arun Hegde, James Oreluk, Michael Frenklach, Andrew K. Packard

Publication date 13 April 2021

Published in SIAM/ASA Journal on Uncertainty Quantification (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1907.00886

zbMATH Keywords

bound-to-bound data collaborationinverse problemmodel discrepancyuncertainty quantification

Mathematics Subject Classification ID

Statistical aspects of big data and data science (62R07) Reasoning under uncertainty in the context of artificial intelligence (68T37)

Abstract: We extended the existing methodology in Bound-to-Bound Data Collaboration (B2BDC), an optimization-based deterministic uncertainty quantification (UQ) framework, to explicitly take into account model discrepancy. The discrepancy was represented as a linear combination of finite basis functions and the feasible set was constructed according to a collection of modified model-data constraints. Formulas for making predictions were also modified to include the model discrepancy function. Prior information about the model discrepancy can be added to the framework as additional constraints. Dataset consistency, a central feature of B2BDC, was generalized based on the extended framework.

Recommendations

• Consistency analysis for massively inconsistent datasets in bound-to-bound data collaboration
• Comparison of statistical and deterministic frameworks of uncertainty quantification
• Data-free inference of the joint distribution of uncertain model parameters
• Quantifying simulator discrepancy in discrete-time dynamical simulators
• When is a model good enough? Deriving the expected value of model improvement via specifying internal model discrepancies

Cites work

• A new look at the statistical model identification
• Bayesian calibration of computer models. (With discussion)
• Combining Field Data and Computer Simulations for Calibration and Prediction
• Comparison of statistical and deterministic frameworks of uncertainty quantification
• Conditional central algorithms for worst case set-membership identification and filtering
• Consistency analysis for massively inconsistent datasets in bound-to-bound data collaboration
• Design and analysis of computer experiments. With comments and a rejoinder by the authors
• Graph implementations for nonsmooth convex programs
• scientific article; zbMATH DE number 1595639 (Why is no real title available?)
• scientific article; zbMATH DE number 1140584 (Why is no real title available?)
• scientific article; zbMATH DE number 1502618 (Why is no real title available?)
• Interpolation of spatial data. Some theory for kriging
• Learning about physical parameters: the importance of model discrepancy
• Numerical approaches for collaborative data processing
• On the value of information in system identification-bounded noise case
• Optimal estimation theory for dynamic systems with set membership uncertainty: An overview
• Parameter tuning for a multi-fidelity dynamical model of the magnetosphere
• Reified Bayesian modelling and inference for physical systems
• Response surface methodology. Process and product optimization using designed experiments
• Using SeDuMi 1.02, A Matlab toolbox for optimization over symmetric cones
• Verification and validation in scientific computing.

Cited in

(1)

• Consistency analysis for massively inconsistent datasets in bound-to-bound data collaboration

Describes a project that uses

Uses Software

• SeDuMi
• CVX
• gss
• GRI-Mech 3.0
• BayesDA
• GitHub

This page was built for publication: Representing model discrepancy in bound-to-bound data collaboration

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5858423)