ANOVA Gaussian process modeling for high-dimensional stochastic computational models
From MaRDI portal
Gaussian processes (60G15) Analysis of variance and covariance (ANOVA) (62J10) Probabilistic models, generic numerical methods in probability and statistics (65C20) PDEs with randomness, stochastic partial differential equations (35R60) Stokes and related (Oseen, etc.) flows (76D07) Stochastic analysis applied to problems in fluid mechanics (76M35)
Abstract: In this paper we present a novel analysis of variance Gaussian process (ANOVA-GP) emulator for models governed by partial differential equations (PDEs) with high-dimensional random inputs. Gaussian process (GP) is a widely used surrogate modeling strategy, but it can become invalid when the inputs are high-dimensional. In this new ANOVA-GP strategy, high-dimensional inputs are decomposed into unions of local low-dimensional inputs, and principal component analysis (PCA) is applied to provide dimension reduction for each ANOVA term. We then systematically build local GP models for PCA coefficients based on ANOVA decomposition to provide an emulator for the overall high-dimensional problem. We present a general mathematical framework of ANOVA-GP, validate its accuracy and demonstrate its efficiency with numerical experiments.
Recommendations
- Towards `\(h\)-\(p\) \textit{adaptive}' generalized ANOVA
- Reduced basis ANOVA methods for partial differential equations with high-dimensional random inputs
- Multiresolution functional ANOVA for large-scale, many-input computer experiments
- Gaussian process emulators for the stochastic finite element method
- An adaptive reduced basis ANOVA method for high-dimensional Bayesian inverse problems
Cites work
- A generalized dimension‐reduction method for multidimensional integration in stochastic mechanics
- A goal-oriented reduced basis methods-accelerated generalized polynomial chaos algorithm
- A posteriori error estimation in finite element analysis
- A survey of projection-based model reduction methods for parametric dynamical systems
- AN ADAPTIVE REDUCED BASIS COLLOCATION METHOD BASED ON PCM ANOVA DECOMPOSITION FOR ANISOTROPIC STOCHASTIC PDES
- Adaptive ANOVA decomposition of stochastic incompressible and compressible flows
- Advanced Lectures on Machine Learning
- An adaptive high-dimensional stochastic model representation technique for the solution of stochastic partial differential equations
- Bayesian calibration of computer models. (With discussion)
- Block-diagonal preconditioning for spectral stochastic finite-element systems
- Comparison between reduced basis and stochastic collocation methods for elliptic problems
- Computer Model Calibration Using High-Dimensional Output
- Efficient input-output model representations
- Error estimates for the ANOVA method with polynomial chaos interpolation: Tensor product functions
- Finite Elements
- Finite elements and fast iterative solvers. With applications in incompressible fluid dynamics
- Gaussian processes for machine learning (GPML) toolbox
- Gaussian processes with built-in dimensionality reduction: applications to high-dimensional uncertainty propagation
- General foundations of high-dimensional model representations
- Generalized principal component analysis
- High dimensional integration of smooth functions over cubes
- High-Order Collocation Methods for Differential Equations with Random Inputs
- IFISS: A Computational Laboratory for Investigating Incompressible Flow Problems
- Is Gauss Quadrature Better than Clenshaw–Curtis?
- Karhunen-Loève approximation of random fields by generalized fast multipole methods
- Kernel principal component analysis for stochastic input model generation
- Manifold learning for the emulation of spatial fields from computational models
- Mean Square Truncation Error in Series Expansions of Random Functions
- Numerical methods for stochastic computations. A spectral method approach.
- On ANOVA expansions and strategies for choosing the anchor point
- On the use of ANOVA expansions in reduced basis methods for parametric partial differential equations
- Reduced Basis Collocation Methods for Partial Differential Equations with Random Coefficients
- Reduced basis ANOVA methods for partial differential equations with high-dimensional random inputs
- Reduced basis techniques for stochastic problems
- Reduced order modeling for nonlinear structural analysis using Gaussian process regression
- Sensitivity analysis using anchored ANOVA expansion and high-order moments computation
- The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations
- The effect of the nugget on Gaussian process emulators of computer models
- Turbulence, Coherent Structures, Dynamical Systems and Symmetry
Cited in
(11)- Towards `\(h\)-\(p\) \textit{adaptive}' generalized ANOVA
- Bayesian multiscale deep generative model for the solution of high-dimensional inverse problems
- ANOVA decomposition of conditional Gaussian processes for sensitivity analysis with dependent inputs
- Data-driven surrogates for high dimensional models using Gaussian process regression on the Grassmann manifold
- An adaptive ANOVA stochastic Galerkin method for partial differential equations with high-dimensional random inputs
- An adaptive Gaussian process emulator for partial differential equations with stochastic parameters
- Gaussian process emulators for the stochastic finite element method
- Multiresolution functional ANOVA for large-scale, many-input computer experiments
- Weak neural variational inference for solving Bayesian inverse problems \textit{without} forward models: applications in elastography
- Full-field order-reduced Gaussian process emulators for nonlinear probabilistic mechanics
- Manifold learning for the emulation of spatial fields from computational models
This page was built for publication: ANOVA Gaussian process modeling for high-dimensional stochastic computational models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q781984)
