Stacking Designs: Designing Multifidelity Computer Experiments with Target Predictive Accuracy

DOI10.1137/22M1532007MaRDI QIDQ6131423zbMATH OpenFDO

Authors

Publication date 5 April 2024

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

Full work available at URL https://arxiv.org/abs/2211.00268

zbMATH Keywords

computer experimentsexperimental designuncertainty quantificationfinite element analysisGaussian process modelingmultilevel modeling

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Sequential statistical design (62L05)

Abstract: In an era where scientific experiments can be very costly, multi-fidelity emulators provide a useful tool for cost-efficient predictive scientific computing. For scientific applications, the experimenter is often limited by a tight computational budget, and thus wishes to (i) maximize predictive power of the multi-fidelity emulator via a careful design of experiments, and (ii) ensure this model achieves a desired error tolerance with some notion of confidence. Existing design methods, however, do not jointly tackle objectives (i) and (ii). We propose a novel stacking design approach that addresses both goals. Using a recently proposed multi-level Gaussian process emulator model, our stacking design provides a sequential approach for designing multi-fidelity runs such that a desired prediction error of $epsilon>0$ is met under regularity assumptions. We then prove a novel cost complexity theorem that, under this multi-level Gaussian process emulator, establishes a bound on the computation cost (for training data simulation) needed to achieve a prediction bound of $epsilon$. This result provides novel insights on conditions under which the proposed multi-fidelity approach improves upon a standard Gaussian process emulator which relies on a single fidelity level. Finally, we demonstrate the effectiveness of stacking designs in a suite of simulation experiments and an application to finite element analysis.

Recommendations

• Sequential design of multi-fidelity computer experiments with effect sparsity
• Budget-limited distribution learning in multifidelity problems
• Conglomerate multi-fidelity Gaussian process modeling, with application to heavy-ion collisions
• Survey of multifidelity methods in uncertainty propagation, inference, and optimization
• Bayesian integrative analysis for multi-fidelity computer experiments

Cites work

• scientific article; zbMATH DE number 4057247 (Why is no real title available?)
• scientific article; zbMATH DE number 45848 (Why is no real title available?)
• scientific article; zbMATH DE number 1215244 (Why is no real title available?)
• scientific article; zbMATH DE number 1967777 (Why is no real title available?)
• scientific article; zbMATH DE number 7370640 (Why is no real title available?)
• scientific article; zbMATH DE number 7306883 (Why is no real title available?)
• scientific article; zbMATH DE number 5681750 (Why is no real title available?)
• Accurate emulators for large-scale computer experiments
• An efficient surrogate model for emulation and physics extraction of large eddy simulations
• Bayesian analysis of hierarchical multifidelity codes
• Convergence of Gaussian process regression with estimated hyper-parameters and applications in Bayesian inverse problems
• Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
• Global Stochastic Optimization with Low-Dispersion Point Sets
• High-dimensional integration: The quasi-Monte Carlo way
• Interpolation of spatial data -- a stochastic or a deterministic problem?
• Interpolation of spatial data. Some theory for kriging
• Minimax and Minimax Projection Designs Using Clustering
• Multilevel Monte Carlo Path Simulation
• Nonlinear information fusion algorithms for data-efficient multi-fidelity modelling
• Numerical methods for stochastic computations. A spectral method approach.
• On Prediction Properties of Kriging: Uniform Error Bounds and Robustness
• On the improved rates of convergence for Matérn-type kernel ridge regression with application to calibration of computer models
• Points on manifolds with asymptotically optimal covering radius
• Predicting the output from a complex computer code when fast approximations are available
• RECURSIVE CO-KRIGING MODEL FOR DESIGN OF COMPUTER EXPERIMENTS WITH MULTIPLE LEVELS OF FIDELITY
• Stochastic processes with sample paths in reproducing kernel Hilbert spaces
• Support points
• The design and analysis of computer experiments
• Verification and validation in scientific computing.

Cited in

(2)

• Conglomerate multi-fidelity Gaussian process modeling, with application to heavy-ion collisions
• Sequential design of multi-fidelity computer experiments with effect sparsity

This page was built for publication: Stacking Designs: Designing Multifidelity Computer Experiments with Target Predictive Accuracy

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6131423)