Low-rank separated representation surrogates of high-dimensional stochastic functions: application in Bayesian inference
From MaRDI portal
(Redirected from Publication:348754)
Abstract: This study introduces a non-intrusive approach in the context of low-rank separated representation to construct a surrogate of high-dimensional stochastic functions, e.g., PDEs/ODEs, in order to decrease the computational cost of Markov Chain Monte Carlo simulations in Bayesian inference. The surrogate model is constructed via a regularized alternative least-square regression with Tikhonov regularization using a roughening matrix computing the gradient of the solution, in conjunction with a perturbation-based error indicator to detect optimal model complexities. The model approximates a vector of a continuous solution at discrete values of a physical variable. The required number of random realizations to achieve a successful approximation linearly depends on the function dimensionality. The computational cost of the model construction is quadratic in the number of random inputs, which potentially tackles the curse of dimensionality in high-dimensional stochastic functions. Furthermore, this vector valued separated representation-based model, in comparison to the available scalar-valued case, leads to a significant reduction in the cost of approximation by an order of magnitude equal to the vector size. The performance of the method is studied through its application to three numerical examples including a 41-dimensional elliptic PDE and a 21-dimensional cavity flow.
Recommendations
- Non-intrusive low-rank separated approximation of high-dimensional stochastic models
- An adaptive reduced basis ANOVA method for high-dimensional Bayesian inverse problems
- Surrogate modeling of high-dimensional problems via data-driven polynomial chaos expansions and sparse partial least square
- Convergence analysis of surrogate-based methods for Bayesian inverse problems
- Non-intrusive tensor reconstruction for high-dimensional random PDEs
Cites work
- scientific article; zbMATH DE number 430432 (Why is no real title available?)
- scientific article; zbMATH DE number 578421 (Why is no real title available?)
- scientific article; zbMATH DE number 2140079 (Why is no real title available?)
- scientific article; zbMATH DE number 840151 (Why is no real title available?)
- scientific article; zbMATH DE number 872525 (Why is no real title available?)
- A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
- A new family of solvers for some classes of multidimensional partial differential equations encountered in kinetic theory modeling of complex fluids
- A stochastic collocation approach to Bayesian inference in inverse problems
- Adaptive proposal distribution for random walk Metropolis algorithm
- Algorithms for Numerical Analysis in High Dimensions
- An adaptive Metropolis algorithm
- An adaptive multi-element generalized polynomial chaos method for stochastic differential equations
- An efficient Bayesian inference approach to inverse problems based on an adaptive sparse grid collocation method
- An introduction to the mathematical theory of inverse problems
- An iterative thresholding algorithm for linear inverse problems with a sparsity constraint
- Bayesian Inference in Econometric Models Using Monte Carlo Integration
- Bayesian analysis of dichotomous quantal response models
- Bayesian inference with optimal maps
- Computational Methods for Inverse Problems
- Convergence Properties of the Nelder--Mead Simplex Method in Low Dimensions
- Delayed rejection in reversible jump Metropolis-Hastings.
- Dimensionality reduction and polynomial chaos acceleration of Bayesian inference in inverse problems
- Discrete inverse problems. Insight and algorithms.
- Efficient Monte Carlo Procedures for Generating Points Uniformly Distributed over Bounded Regions
- Equation of state calculations by fast computing machines
- Inverse Problem Theory and Methods for Model Parameter Estimation
- Learning to Predict Physical Properties using Sums of Separable Functions
- Linear models. Least squares and alternatives.
- Markov Chain Monte Carlo Methods for High Dimensional Inversion in Remote Sensing
- Markov Chains
- Markov chains for exploring posterior distributions. (With discussion)
- Multiparameter Univariate Bayesian Analysis
- Multivariate regression and machine learning with sums of separable functions
- Non-intrusive low-rank separated approximation of high-dimensional stochastic models
- Nonstationary inverse problems and state estimation
- Numerical operator calculus in higher dimensions
- Ordering and improving the performance of Monte Carlo Markov chains.
- Principal component analysis of three-mode data by means of alternating least squares algorithms
- Spectral Methods for Uncertainty Quantification
- Stochastic model reduction for chaos representations
- Tensor Decompositions and Applications
- When did Bayesian inference become ``Bayesian?
Cited in
(7)- Stochastic collocation algorithms using \(l_1\)-minimization for Bayesian solution of inverse problems
- Sparse low-rank separated representation models for learning from data
- Polynomial meta-models with canonical low-rank approximations: numerical insights and comparison to sparse polynomial chaos expansions
- The optimization landscape for fitting a rank-2 tensor with a rank-1 tensor
- Surrogate modeling of high-dimensional problems via data-driven polynomial chaos expansions and sparse partial least square
- scientific article; zbMATH DE number 6975079 (Why is no real title available?)
- Non-intrusive low-rank separated approximation of high-dimensional stochastic models
This page was built for publication: Low-rank separated representation surrogates of high-dimensional stochastic functions: application in Bayesian inference
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q348754)
