Alphabeta Math
LemmaStatement: AI-adaptedProof: AI-generatedPipeline-generatedprecheck passverified 2026-10-22 (gpt-6.1-sol)
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.

Tensor products of tempered distributions

Statement

Let p,q≥1, let u∈S′(Rp) and v∈S′(Rq), and use the Schwartz seminorms pαβ(f)=sup⁡x∣xα∂βf(x)∣. The iterated pairing

w(Φ):=u ⁣(x↦v ⁣(y↦Φ(x,y))),Φ∈S(Rp+q),

defines a tempered distribution w∈S′(Rp+q). It satisfies w(φ⊗ψ)=u(φ)v(ψ) for Schwartz functions φ,ψ. It agrees with the distribution tensor product on compactly supported tests, is unique with these product values, and its finite iterates are associative.

Facts & Assumptions

Given: Integers p,q≥1, u∈S′(Rp) and v∈S′(Rq).

[F1]

Schwartz space and its seminorms: the seminorms pαβ define the Schwartz space.

[F2]

Schwartz topology and convergence: a neighbourhood of zero restricts only finitely many seminorms.

[F3]

Tempered distribution: u and v are continuous complex-linear functionals on their Schwartz spaces.

[F4]

Test function cutoffs and euclidean localization: a compact set inside an open set admits a smooth compactly supported cutoff equal to one near that compact set.

[F5]

Finite sums of product tests are dense on product open sets: finite sums of compactly supported product tests are dense in the compactly supported smooth test topology, with approximating supports in a fixed compact product.

[F6]

Tensor product distributions and iterated pairings: on compactly supported tests the distribution tensor product is the iterated pairing, has the product values, is unique, and iterates associatively.

Proof

Given: [F1]–[F6].

1.1F1F2F3givenalgebra

By continuity [F3], choose a basic neighbourhood Vu={f:pαjβj(f)<εj, 1≤j≤k} on which ∣u(f)∣<1. Intersecting with p00(f)<1 if necessary ensures k≥1. Set Mu=max⁡1≤j≤k{∣αj∣,∣βj∣}, PMu(f)=max⁡∣α∣,∣β∣≤Mupαβ(f), and cu=2max⁡jεj−1. For D(f)=max⁡j(pαjβj(f)/εj)>0, the test f/(2D(f)) lies in Vu, so ∣u(f)∣≤2D(f)≤cuPMu(f). If D(f)=0, every positive scalar multiple of f lies in Vu, forcing u(f)=0 and the same bound. Applying this argument to v gives finite Mv,cv with ∣v(g)∣≤cvPMv(g).

2.1F1F2F3step 1.1given

For fixed Φ∈S(Rp+q), the map x↦Φ(x,⋅) is smooth with values in S(Rq): Taylor's formula in the x coordinates, with its remainder differentiated in y, tends to zero in each pγδ on every compact x set because every mixed Schwartz seminorm of Φ is finite. Continuity of v therefore makes h(x):=v(Φ(x,⋅)) smooth and gives ∂xβh(x)=v(∂xβΦ(x,⋅)). For every α,β, [F3] and step 1.1 yield sup⁡x∣xα∂xβh(x)∣≤cvmax⁡∣γ∣,∣δ∣≤Mvsup⁡x,y∣xαyγ∂xβ∂yδΦ(x,y)∣<∞. Thus h∈S(Rp).

3.1F1F3F6step 1.1step 2.1givenalgebra

Define w(Φ)=u(h). Applying the bound in step 1.1 to the finitely many x seminorms with indices at most Mu, and then the displayed estimate in step 2.1, bounds ∣w(Φ)∣ by a finite constant times the maximum of finitely many product Schwartz seminorms of Φ. Hence w is continuous on S(Rp+q). For Φ=φ⊗ψ, the inner pairing is φ(x)v(ψ), giving the claimed product value; on compactly supported tests the formula is the same iterated pairing as [F6].

4.1F1F4F5step 3.1∎

Finite sums of Schwartz product tests are dense in S(Rp+q). Indeed, choose χ∈Cc∞(Rp) and η∈Cc∞(Rq) equal to one near the unit balls by [F4], and set χR(x)=χ(x/R) and ηR(y)=η(y/R). The product rule and rapid decay show χR(x)ηR(y)Φ(x,y)→Φ(x,y) in every product Schwartz seminorm: the undifferentiated tails lie outside balls of radius proportional to R, and each differentiated cutoff contributes a factor R−j on the same tails. For fixed R, [F5] approximates this compactly supported smooth function by finite sums of product tests with common compact support; on that support, uniform convergence of all derivatives implies convergence in every Schwartz seminorm. Thus any two continuous functionals with the same product values agree on a dense subspace and hence everywhere. The same density argument, iterated over finitely many factors, shows that different parenthesizations of the constructed tensor product agree; this also agrees with the compact-test associativity in [F6].

Depends on

Dependency tree · 2 levels

Sources