Alphabeta Math
CorollaryStatement: AI-adaptedProof: AI-generatedPipeline-generatedjudge 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.

Unaliased fixed-grid spectral bias under the exact no-alias condition

Statement

Assume the model and AC/DC of Spectral alias identity for fixed regular-grid sampling. Fix ω0∈Rd, and put λ0=δω0 modulo 2π. Suppose the pointwise periodized sum gδ has a bounded torus representative continuous at λ0, agreeing with that sum at λ0. Use the kernel and bandwidths of Vanishing bias of a smoothed regular-grid spatial periodogram. Define the physically normalized estimates at ω0 by f^nraw(ω0)=δdIn(δω0),f^nsm(ω0)=δdf~n,bn(δω0). Then the smoother is defined eventually and both expectations converge to aδ(ω0):=∑k∈Zdf(ω0+2πk/δ). In particular, both have vanishing bias for the unaliased value f(ω0) if and only if f(ω0+2πk/δ)=0for every k≠0. This exact pointwise condition does not restrict the general aliased-target theorem. A sufficient global condition is that f is continuous and supported in a compact subset of the open Nyquist cube (−π/δ,π/δ)d, with ω0 in that cube. No variance, consistency or CLT conclusion is asserted.

Facts & Assumptions

Given: The fixed-spacing continuous-space model, a bounded periodized density representative continuous at the target and agreeing there with its sum, the kernel/bandwidth assumptions, and AC/DC.

[F1]

Spectral alias identity for fixed regular-grid sampling: gδ is the torus density of the sampled field and its pointwise sum is δ−d∑kf((λ+2πk)/δ).

[F2]

Vanishing bias of a smoothed regular-grid spatial periodogram: for a bounded torus density continuous at the chosen target, both raw and smoothed periodogram expectations converge to that density value; smoothing denominators are eventually positive.

[F3]

Regular-grid spatial periodogram and smoothed estimator: the normalized Fourier-grid smoother is the specified finite nonnegative weighted sum of angular periodogram ordinates.

[A1]

The Axiom of Choice and The axiom of dependent choice: a relation in which every element is related to something admits an N-indexed chain: AC and DC are inherited through the lattice spectral representation and its bias theorem.

Proof

1.1A1F1F2F3givenalgebra

By [F1] the sampled field has torus density gδ, and the assumed representative meets all hypotheses of [F2]. Thus the denominator in [F3] is eventually positive and both angular expectations tend to gδ(λ0). Multiplying by δd and reindexing the convergent nonnegative sum at the target gives δdgδ(λ0)=∑kf(ω0+2πk/δ)=aδ(ω0). This proves the general aliased limit for both physically normalized estimates.

2.1F1step 1.1givenalgebra∎

Every alias summand is nonnegative. Therefore aδ(ω0)−f(ω0)=∑k≠0f(ω0+2πk/δ) is zero if and only if every summand is zero. This proves both directions of the stated exact condition. Under the sufficient global condition, only the central translate is nonzero for frequencies in the Nyquist cube; its support stays away from the boundary. Hence the periodic density is bounded and continuous across the boundary as well, and has no noncentral alias at the stated target. All hypotheses and the exact condition are then verified.

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

45 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