Adaptive sampling-based quadrature rules for efficient Bayesian prediction
From MaRDI portal
Publication:782004
DOI10.1016/J.JCP.2020.109537zbMATH Open1437.76043arXiv1907.08418OpenAlexW3025665987MaRDI QIDQ782004FDOQ782004
L. M. M. van den Bos, W. A. A. M. Bierbooms, B. Sanderse
Publication date: 21 July 2020
Published in: Journal of Computational Physics (Search for Journal in Brave)
Abstract: A novel method is proposed to infer Bayesian predictions of computationally expensive models. The method is based on the construction of quadrature rules, which are well-suited for approximating the weighted integrals occurring in Bayesian prediction. The novel idea is to construct a sequence of nested quadrature rules with positive weights that converge to a quadrature rule that is weighted with respect to the posterior. The quadrature rules are constructed using a proposal distribution that is determined by means of nearest neighbor interpolation of all available evaluations of the posterior. It is demonstrated both theoretically and numerically that this approach yields accurate estimates of the integrals involved in Bayesian prediction. The applicability of the approach for a fluid dynamics test case is demonstrated by inferring accurate predictions of the transonic flow over the RAE2822 airfoil with a small number of model evaluations. Here, the closure coefficients of the Spalart-Allmaras turbulence model are considered to be uncertain and are calibrated using wind tunnel measurements.
Full work available at URL: https://arxiv.org/abs/1907.08418
Bayesian inference (62F15) Probabilistic models, generic numerical methods in probability and statistics (65C20) Numerical quadrature and cubature formulas (65D32) Stochastic analysis applied to problems in fluid mechanics (76M35) Transonic flows (76H05)
Cites Work
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Monotone Piecewise Cubic Interpolation
- Bayesian calibration of computer models. (With discussion)
- VPS: VORONOI PIECEWISE SURROGATE MODELS FOR HIGH-DIMENSIONAL DATA FITTING
- Monte Carlo sampling methods using Markov chains and their applications
- Equation of State Calculations by Fast Computing Machines
- Random variate generation by numerical inversion when only the density is known
- Lebesgue functions and Lebesgue constants in polynomial interpolation
- Calculation of Gauss Quadrature Rules
- Spectral Methods for Uncertainty Quantification
- Scattered Data Interpolation: Tests of Some Method
- Adaptive Leja Sparse Grid Constructions for Stochastic Collocation and High-Dimensional Approximation
- Construction of positive definite cubature formulae and approximation of functions via Voronoi tessellations
- Centroidal Voronoi Tessellations: Applications and Algorithms
- Quadrature rules and distribution of points on manifolds
- Sparse, adaptive Smolyak quadratures for Bayesian inverse problems
- Simple cubature formulas with high polynomial exactness
- Simplex-stochastic collocation method with improved scalability
- Adaptive construction of surrogates for the Bayesian solution of inverse problems
- Sparse deterministic approximation of Bayesian inverse problems
- Simplex stochastic collocation with random sampling and extrapolation for nonhypercube probability spaces
- Bayesian estimates of parameter variability in the \(k-\varepsilon\) turbulence model
- Predictive RANS simulations via Bayesian model-scenario averaging
- Lectures on the nearest neighbor method
- A continuation multi level Monte Carlo (C-MLMC) method for uncertainty quantification in compressible inviscid aerodynamics
- Non-intrusive uncertainty quantification using reduced cubature rules
- Introduction to Uncertainty Quantification
- Solution of inverse problems with limited forward solver evaluations: a Bayesian perspective
- Transformed density rejection with inflection points
- Generation and application of multivariate polynomial quadrature rules
- A transport-based multifidelity preconditioner for Markov chain Monte Carlo
- Convergence of Probability Densities Using Approximate Models for Forward and Inverse Problems in Uncertainty Quantification
- Posterior consistency for Gaussian process approximations of Bayesian posterior distributions
- Convergence analysis of surrogate-based methods for Bayesian inverse problems
- Generating Nested Quadrature Rules with Positive Weights based on Arbitrary Sample Sets
- Bayesian Model Calibration with Interpolating Polynomials based on Adaptively Weighted Leja Nodes
- A Measure-Theoretic Interpretation of Sample Based Numerical Integration with Applications to Inverse and Prediction Problems under Uncertainty
- Adaptive Bayesian Inference for Discontinuous Inverse Problems, Application to Hyperbolic Conservation Laws
Cited In (3)
Uses Software
This page was built for publication: Adaptive sampling-based quadrature rules for efficient Bayesian prediction
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q782004)
