English

Efficient multivariate approximation on the cube

Numerical Analysis 2021-01-22 v3 Numerical Analysis

Abstract

We combine a periodization strategy for weighted L2L_{2}-integrands with efficient approximation methods in order to approximate multivariate non-periodic functions on the high-dimensional cube [12,12]d\left[-\frac{1}{2},\frac{1}{2}\right]^{d}. Our concept allows to determine conditions on the dd-variate torus-to-cube transformations ψ:[12,12]d[12,12]d{\psi:\left[-\frac{1}{2},\frac{1}{2}\right]^{d}\to\left[-\frac{1}{2},\frac{1}{2}\right]^{d}} such that a non-periodic function is transformed into a smooth function in the Sobolev space Hm(Td)\mathcal H^{m}(\mathbb{T}^{d}) when applying ψ\psi. We adapt some L(Td)L_{\infty}(\mathbb{T}^{d})- and L2(Td)L_{2}(\mathbb{T}^{d})-approximation error estimates for single rank-11 lattice approximation methods and adjust algorithms for the fast evaluation and fast reconstruction of multivariate trigonometric polynomials on the torus in order to apply these methods to the non-periodic setting. We illustrate the theoretical findings by means of numerical tests in up to d=5d=5 dimensions.

Keywords

Cite

@article{arxiv.1912.03090,
  title  = {Efficient multivariate approximation on the cube},
  author = {Robert Nasdala and Daniel Potts},
  journal= {arXiv preprint arXiv:1912.03090},
  year   = {2021}
}

Comments

arXiv admin note: substantial text overlap with arXiv:1805.09106