Stability of discrete empirical interpolation and gappy proper orthogonal decomposition with randomized and deterministic sampling points

DOI10.1137/19M1307391zbMATH Open1453.65287arXiv1808.10473OpenAlexW3089316803MaRDI QIDQ5131990FDOQ5131990

Authors: Benjamin Peherstorfer, Zlatko Drmač, Serkan Gugercin 

Publication date: 9 November 2020

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Abstract: This work investigates the stability of (discrete) empirical interpolation for nonlinear model reduction and state field approximation from measurements. Empirical interpolation derives approximations from a few samples (measurements) via interpolation in low-dimensional spaces. It has been observed that empirical interpolation can become unstable if the samples are perturbed due to, e.g., noise, turbulence, and numerical inaccuracies. The main contribution of this work is a probabilistic analysis that shows that stable approximations are obtained if samples are randomized and if more samples than dimensions of the low-dimensional spaces are used. Oversampling, i.e., taking more sampling points than dimensions of the low-dimensional spaces, leads to approximations via regression and is known under the name of gappy proper orthogonal decomposition. Building on the insights of the probabilistic analysis, a deterministic sampling strategy is presented that aims to achieve lower approximation errors with fewer points than randomized sampling by taking information about the low-dimensional spaces into account. Numerical results of reconstructing velocity fields from noisy measurements of combustion processes and model reduction in the presence of noise demonstrate the instability of empirical interpolation and the stability of gappy proper orthogonal decomposition with oversampling.

Full work available at URL: https://arxiv.org/abs/1808.10473

Recommendations

• The generalized empirical interpolation method: stability theory on Hilbert spaces with an application to the Stokes equation
• Nonlinear model reduction via discrete empirical interpolation
• A New Look at Proper Orthogonal Decomposition
• Unsteady flow sensing and estimation via the gappy proper orthogonal decomposition
• An `empirical interpolation' method: Application to efficient reduced-basis discretization of partial differential equations

zbMATH Keywords

model reductionnoisy observationsnonlinear model reductionempirical interpolationgappy proper orthogonal decompositionrandomized model reduction

Mathematics Subject Classification ID

Numerical interpolation (65D05) Orthogonalization in numerical linear algebra (65F25) Numerical methods for partial differential equations, initial value and time-dependent initial-boundary value problems (65M99) Numerical solution of discretized equations for initial value and initial-boundary value problems involving PDEs (65M22)

Cites Work

• Exact matrix completion via convex optimization
• An `empirical interpolation' method: Application to efficient reduced-basis discretization of partial differential equations
• A generalized empirical interpolation method: application of reduced basis techniques to data assimilation
• Nonlinear model reduction via discrete empirical interpolation
• Missing Point Estimation in Models Described by Proper Orthogonal Decomposition
• Unsteady flow sensing and estimation via the gappy proper orthogonal decomposition
• A survey of projection-based model reduction methods for parametric dynamical systems
• Efficient non linear model reduction via a least-squares Petrov-Galerkin projection and compressive tensor approximations
• Efficient reduced-basis treatment of nonaffine and nonlinear partial differential equations
• The GNAT method for nonlinear model reduction: effective implementation and application to computational fluid dynamics and turbulent flows
• On the stability and accuracy of least squares approximations
• An accelerated greedy missing point estimation procedure
• Title not available (Why is that?)
• Parameter multi-domain `\(hp\)' empirical interpolation
• Simple models of turbulent flows
• Refined perturbation bounds for eigenvalues of Hermitian and non-Hermitian matrices
• Discrete least squares polynomial approximation with random evaluations - application to parametric and stochastic elliptic PDEs
• POD-DEIM model order reduction for strain-softening viscoplasticity
• Online adaptive model reduction for nonlinear systems via low-rank updates
• Survey of multifidelity methods in uncertainty propagation, inference, and optimization
• Convergence estimates in probability and in expectation for discrete least squares with noisy evaluations at random points
• Localized Discrete Empirical Interpolation Method
• Greedy Algorithms for Optimal Measurements Selection in State Estimation Using Reduced Models
• Sensor placement in nuclear reactors based on the generalized empirical interpolation method
• Stabilization of (G)EIM in presence of measurement noise: application to nuclear reactor physics
• The generalized empirical interpolation method: stability theory on Hilbert spaces with an application to the Stokes equation
• A new selection operator for the discrete empirical interpolation method -- improved a priori error bound and extensions
• Geometric subspace updates with applications to online adaptive nonlinear model reduction
• The Discrete Empirical Interpolation Method: Canonical Structure and Formulation in Weighted Inner Product Spaces

Cited In (30)

• An empirical interpolation method for two-dimensional vector fields and vector measurements
• Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data
• An adaptive, training-free reduced-order model for convection-dominated problems based on hybrid snapshots
• S-OPT: a points selection algorithm for hyper-reduction in reduced order models
• Signal generator framework-based variation expansion-matching method for polynomial systems
• Model reduction for multi-scale transport problems using model-form preserving least-squares projections with variable transformation
• Randomized greedy magic point selection schemes for nonlinear model reduction
• Learning physics-based models from data: perspectives from inverse problems and model reduction
• Efficient and accurate nonlinear model reduction via first-order empirical interpolation
• Randomized GCUR decompositions
• Active Operator Inference for Learning Low-Dimensional Dynamical-System Models from Noisy Data
• Reduced-order modeling of nonlinear structural dynamical systems via element-wise stiffness evaluation procedure combined with hyper-reduction
• Stabilization of (G)EIM in presence of measurement noise: application to nuclear reactor physics
• A registration method for reduced basis problems using linear optimal transport
• The generalized empirical interpolation method: stability theory on Hilbert spaces with an application to the Stokes equation
• Energy-conserving hyper-reduction and temporal localization for reduced order models of the incompressible Navier-Stokes equations
• Reduced order model of standard \(k\)-\(\epsilon\) turbulence model
• A training set subsampling strategy for the reduced basis method
• An extended DEIM algorithm for subset selection and class identification
• The discrete empirical interpolation method in class identification and data summarization
• The Discrete Empirical Interpolation Method: Canonical Structure and Formulation in Weighted Inner Product Spaces
• Model order reduction strategies for weakly dispersive waves
• Predictive reduced order modeling of chaotic multi-scale problems using adaptively sampled projections
• Cross interpolation for solving high-dimensional dynamical systems on low-rank Tucker and tensor train manifolds
• Sparse discrete empirical interpolation method: state estimation from few sensors
• Collocation methods for nonlinear differential equations on low-rank manifolds
• Adaptive sparse interpolation for accelerating nonlinear stochastic reduced-order modeling with time-dependent bases
• Front transport reduction for complex moving fronts
• Stabilized reduced-order models for unsteady incompressible flows in three-dimensional parametrized domains
• A new selection operator for the discrete empirical interpolation method -- improved a priori error bound and extensions

This page was built for publication: Stability of discrete empirical interpolation and gappy proper orthogonal decomposition with randomized and deterministic sampling points

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5131990)