Coherence Motivated Sampling and Convergence Analysis of Least-Squares Polynomial Chaos Regression

Abstract

Independent sampling of orthogonal polynomial bases via Monte Carlo is of interest for uncertainty quantification of models, using Polynomial Chaos (PC) expansions. It is known that bounding the spectral radius of a random matrix consisting of PC samples, yields a bound on the number of samples necessary to identify coefficients in the PC expansion via solution to a least-squares regression problem. We present a related analysis which guarantees a mean square convergence using a coherence parameter of the sampled PC basis that may be both analytically bounded and computationally estimated. Utilizing asymptotic results for orthogonal polynomials, we bound the coherence parameter for polynomials of Hermite and Legendre type under each respective natural sampling distribution. In both polynomial bases we identify an importance sampling distribution which yields a bound with weaker dependence on the order of the PC basis. For more general orthonormal bases, we propose the coherence-optimal sampling: a Markov Chain Monte Carlo sampling, which directly uses the basis functions under consideration to achieve a statistical optimality among all such sampling schemes with identical support, and which guarantees recovery with a number of samples that is, up to logarithmic factors, linear in the number of basis functions considered. We demonstrate these different sampling strategies numerically in both high-order and high-dimensional manufactured PC expansions. In addition, the quality of each sampling method is compared in the identification of solutions to two differential equations, one with a high-dimensional random input and the other with a high-order PC expansion. In all observed cases the coherence-optimal sampling leads to similar or considerably improved accuracy over the other considered sampling distributions.

arXiv commentarXiv admin note: substantial text overlap with arXiv:1408.4157

References in corpus (6)

• User-friendly tail bounds for sums of random matrices
• A non-adapted sparse approximation of PDEs with stochastic inputs
• Compressive Sampling of Polynomial Chaos Expansions: Convergence Analysis and Sampling Strategies
• A weighted L1-minimization approach for sparse polynomial chaos expansions
• Coarse Stability and Bifurcation Analysis Using Stochastic Simulators: Kinetic Monte Carlo Examples
• On discrete least square projection in unbounded domain with random evaluations and its application to parametric uncertainty quantification

Cited by in corpus (35)

• Sparse Polynomial Chaos Expansions: Literature Survey and Benchmark
• Least Squares Polynomial Chaos Expansion: A Review of Sampling Strategies
• Sparse Polynomial Chaos Expansions via Compressed Sensing and D-optimal Design
• Sparse Identification of Nonlinear Dynamical Systems via Reweighted ${\ell }_1$-regularized Least Squares
• On Polynomial Chaos Expansion via Gradient-enhanced ${\ell }_1$-minimization
• A Christoffel function weighted least squares algorithm for collocation approximations
• Data-driven polynomial ridge approximation using variable projection
• Basis Adaptive Sample Efficient Polynomial Chaos (BASE-PC)
• A gradient enhanced ${\ell }_1$-minimization for sparse approximation of polynomial chaos expansions
• Optimal pointwise sampling for $L^2$ approximation
• Regression-based sparse polynomial chaos for uncertainty quantification of subsurface flow models
• Multifidelity uncertainty quantification with models based on dissimilar parameters
• Randomized Numerical Linear Algebra: Foundations & Algorithms
• Bi-fidelity Variational Auto-encoder for Uncertainty Quantification
• Boosted optimal weighted least-squares
• Sensitivity-enhanced generalized polynomial chaos for efficient uncertainty quantification
• Robust Adaptive Least Squares Polynomial Chaos Expansions in High-Frequency Applications
• Multi-fidelity Machine Learning for Uncertainty Quantification and Optimization
• A preconditioning approach for improved estimation of sparse polynomial chaos expansions
• Stable approximation of Helmholtz solutions in the disk by evanescent plane waves
• Orbit Uncertainty Propagation and Sensitivity Analysis With Separated Representations
• Power Module Heat Sink Design Optimization with Ensembles of Data-Driven Polynomial Chaos Surrogate Models
• Fourier Sparse Leverage Scores and Approximate Kernel Learning
• On the Sampling Problem for Kernel Quadrature
• Approximate and integrate: Variance reduction in Monte Carlo integration via function approximation
• Calibrating a Finite-strain Phase-field Model of Fracture for Bonded Granular Materials with Uncertainty Quantification
• Parametric/Stochastic Model Reduction: Low-Rank Representation, Non-Intrusive Bi-Fidelity Approximation, and Convergence Analysis
• Noise-robust multi-fidelity surrogate modelling for parametric partial differential equations
• Quadrature Sampling of Parametric Models with Bi-fidelity Boosting
• Signal reconstruction using determinantal sampling
• Optimal sampling and Christoffel functions on general domains
• Stable approximation of Helmholtz solutions in the 3D ball using evanescent plane waves
• GenMod: A generative modeling approach for spectral representation of PDEs with random inputs
• Weighted approximate Fekete points: Sampling for least-squares polynomial approximation
• A sampling method based on highest density regions: Applications to surrogate models