Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck passjudge pass (gpt-6-luna)
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

The Matérn native space has its weighted Fourier norm and strict Gram positivity

Statement

Assume the Axiom of Choice conservatively for the declared L2 Hilbert-space dependency. Fix d≥1, ν>0, a>0, and s=ν+d/2, and let k,w be the normalized Matérn kernel and positive spectral density of The normalized Matérn kernel in the 2π Fourier convention. Let Hk(Rd) consist of the inverse Fourier representatives Ug with ∫∣g(ξ)∣2/w(ξ) dξ<∞, with inner product ⟨Ug,Uh⟩k=∫Rdg(ξ)h(ξ)‾w(ξ)dξ. Then this is a Hilbert space, each kernel section k(⋅−y) reproduces evaluation at y, and its span is dense. There are constants 0<c≤C<∞, depending only on d,ν,a, such that c∥U∥Hs(Rd)≤∥U∥k≤C∥U∥Hs(Rd). For every finite list of distinct sites X=(x1,…,xN), the Matérn Gram matrix KX=(k(xi−xj)) is strictly positive definite.

Facts & Assumptions

Given: The Axiom of Choice, fixed d,ν,a, the definitions of k,w,Hs, and distinct sites X.

[F1]

w(ξ)=A(a2+4π2∣ξ∣2)−s is positive and integrable with integral one; the inverse Fourier representation of Hs is injective. (The normalized Matérn kernel in the 2π Fourier convention, The Matérn spectral integral, normalization, and radial formula, The weighted Fourier model is an injective Hilbert representation)

[F2]

The Hs Fourier norm is ∫(1+∣ξ∣2)s∣g(ξ)∣2dξ, and the inverse Fourier representative is injective. (Fourier Bessel-potential and restriction-quotient Sobolev spaces, The weighted Fourier model is an injective Hilbert representation)

[F3]

A closed subspace of a Hilbert space has an orthogonal complement, and a subspace whose orthogonal complement is zero is dense. (Orthogonal decomposition by a closed subspace)

[F4]

A nonzero univariate polynomial has at most its degree many roots. The real numbers form an ordered field, hence are infinite. (A nonzero polynomial of degree n over an integral domain has at most n distinct roots, The reals form a totally ordered field, The reals form a field)

[F6]

A nonnegative measurable function with integral zero vanishes almost everywhere. (A nonnegative measurable function has integral 0 exactly when it vanishes almost everywhere)

[A1]

Full AC is carried only to conservatively cover the legacy declared dependency of the published L2 completeness result; the Fourier calculations and finite-site selection use no arbitrary-index choice. (The Axiom of Choice, The Axiom of Countable Choice (ACω))

Proof technique: direct.

Proof

technique · direct

Given: The data above; all Fourier transforms use the negative-sign 2π convention.

1.1F1F2F7A1algebra

Since w=A(a2+4π2∣ξ∣2)−s, the two positive functions (a2+4π2∣ξ∣2)s and (1+∣ξ∣2)s have ratio bounded above and below by positive constants: compare a2+4π2r2 with 1+r2 separately using the minimum and maximum of a2 and 4π2, and use monotonicity of the positive power [F7]. Multiplying by A−1 and taking square roots gives the displayed norm equivalence. Thus the weighted L2 norms defining Hk and Hs are equivalent and have the same representatives by injectivity [F1,F2]; completeness of Hs implies completeness of Hk.

1.2F1F5F6algebra

For a complex coefficient vector c=(cj), the Hermitian quadratic form of the Gram matrix is c∗KXc=∫Rdw(ξ)∣∑j=1Ncje−2πixj⋅ξ∣2dξ. If it is zero, [F6] says w times the squared modulus of the continuous exponential sum vanishes almost everywhere. Since w>0, the sum is zero almost everywhere. A continuous function zero almost everywhere is zero everywhere, because any point of nonzero value would have a neighborhood on which its modulus is bounded below, and such a neighborhood contains a positive-measure box [F5].

2.1F1F5step 1.1algebra

The spectral representative of k(⋅−y) is w(ξ)e−2πiy⋅ξ. Its Hk norm squared is ∫w=1, and for Ug∈Hk, ⟨Ug,k(⋅−y)⟩k=∫g(ξ)e2πiy⋅ξdξ=Ug(y). The integral exists because Cauchy--Schwarz bounds ∫∣g∣ by ∥g∥k(∫w)1/2. Hence point evaluation is continuous and the stated kernel sections reproduce it.

3.1F1F3step 2.1

If U is orthogonal to every section centered at y∈Rd, step 2.1 gives U(y)=0 for every y. Fourier injectivity [F1] then makes U=0 in Hk. By [F3] the span of all sections is dense.

4.1F4F5step 1.2choosealgebra∎

Choose v=(1,t,…,td−1) so that the numbers v⋅xj are pairwise distinct. For each pair i≠j, the polynomial (xi−xj)⋅(1,t,…,td−1) is nonzero because xi−xj≠0; [F4] says its zero set is finite. The ordered field R has 0<1, so the natural multiples 0,1,1+1,… are strictly increasing and R is infinite. Thus a real t exists outside the finite union of root sets. Restrict the vanishing exponential sum to ξ=uv and differentiate at u=0 for orders r=0,…,N−1. This gives ∑jcj(v⋅xj)r=0. For each j, the Lagrange polynomial Lj(t)=∏i≠jt−v⋅xiv⋅xj−v⋅xi has degree at most N−1 and satisfies Lj(v⋅xi)=1 if i=j and 0 otherwise. Taking the same linear combination of the moment equations as the coefficients of Lj yields cj=0. Thus all cj=0, and the real Gram matrix is strictly positive definite.

Depends on

Used by

Dependency tree · two levels

99 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources