A characterization of Sobolev spaces on the sphere and an extension of Stolarsky's invariance principle to arbitrary smoothness

Abstract

In this paper we study reproducing kernel Hilbert spaces of arbitrary smoothness on the sphere ${\mathbb{S}}^d\subset {\mathbb{R}}^{d+1}$. The reproducing kernel is given by an integral representation using the truncated power function $(\mathbf{x}\cdot \mathbf{z}-t{)}_{+}^{\beta -1}$ defined on spherical caps centered at $\mathbf{z}$ of height $t$, which reduce to an integral over indicator functions of spherical caps as studied in [J. Brauchart, J. Dick, arXiv:1101.4448v1 [math.NA], to appear in Proc. Amer. Math. Soc.] for $\beta =1$. This is in analogy to the generalization of the reproducing kernel to arbitrary smoothness on the unit cube. We show that the reproducing kernel is a sum of a Kamp{é} de F{é}riet function and the Euclidean distance $\Vert \mathbf{x}-\mathbf{y}\Vert$ of the arguments of the kernel raised to the power of $2\beta -1$ if $2\beta -1$ is not an even integer; otherwise the logarithm of the distance $\Vert \mathbf{x}-\mathbf{y}\Vert$ appears. For $\beta \in \mathbb{N}$ the Kampé de Fériet function reduces to a polynomial, giving a simple closed form expression for the reproducing kernel. Using this space we can generalize Stolarsky's invariance principle to arbitrary smoothness. Previously, Warnock's formula, which is the analogue to Stolarsky's invariance principle for the unit cube $[0,1{]}^s$, has been generalized using similar techniques [J. Dick, Ann. Mat. Pura. Appl., (4) 187 (2008), no. 3, 385--403].

References in corpus (2)

• QMC designs: optimal order Quasi Monte Carlo Integration schemes on the sphere
• A simple Proof of Stolarsky's Invariance Principle

Cited by in corpus (12)

• Distributing many points on spheres: minimal energy and designs
• Infinite-dimensional inverse problems with finite measurements
• Hyperuniform point sets on the sphere: deterministic constructions
• Regularization Matters: A Nonparametric Perspective on Overparametrized Neural Network
• Discrepancy and numerical integration on metric measure spaces
• Spatial low-discrepancy sequences, spherical cone discrepancy, and applications in financial modeling
• Discrepancy estimates for variance bounding Markov chain quasi-Monte Carlo
• "Magic" numbers in Smale's 7th problem
• A Comparison of Popular Point Configurations on ${\mathbb{S}}^2$
• Discrepancy bounds for uniformly ergodic Markov chain quasi-Monte Carlo
• Posterior Integration on a Riemannian Manifold
• Function recovery on manifolds using scattered data