Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedjudge pass (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.

Positive definite functions give positive bounded functionals on the transform core

Statement

Assume the Axiom of Choice and Dependent Choice. Let G be a locally compact Hausdorff abelian group with Haar measure mG and dual G^, and let ϕ:G→C be continuous and positive definite, with k:=ϕ(0)≥0. Define Lϕ(f):=∫Gf(x) ϕ(−x) dmG(x),f∈L1(G,mG). Then Lϕ is a well-defined linear functional with ∣Lϕ(f)∣≤k∥f∥1, the integrated positivity Lϕ(f∗f∗) ≥ 0(f∈L1(G,mG)) holds, and the Cauchy-Schwarz-type bound ∣Lϕ(f)∣2≤k Lϕ(f∗f∗)≤k2 ∥f^∥∞2 holds. Consequently Lϕ vanishes on {f:f^=0} and descends to a positive linear functional Fϕ on the transform core {f^:f∈L1(G,mG)}⊆C0(G^) with ∣Fϕ(h)∣≤k ∥h∥∞. No condition on the growth or integrability of ϕ beyond continuity and positive definiteness is needed.

Facts & Assumptions

Given: The Axiom of Choice and Dependent Choice, a locally compact Hausdorff abelian group G with Haar measure mG, a continuous positive definite ϕ with k=ϕ(0), the functional Lϕ, and A=L1(G,mG).

[F1]

Positive definiteness gives ϕ(−x)=ϕ(x)‾ and ∣ϕ(x)∣≤ϕ(0)=k, and ϕ is uniformly continuous on compact sets (Positive definite functions on an abelian group).

[F2]

Convolution and involution make A a commutative Banach ∗-algebra with ∥g∗h∥1≤∥g∥1∥h∥1, (g∗h)∗=h∗∗g∗ and g∗∗=g (L^1 of an LCA group is a commutative Banach star algebra under convolution); the transform satisfies g∗h^=g^h^ and g∗^=g^‾ (Fourier transform intertwines translation, modulation and convolution).

[F3]

The approximate identity {uU}⊆Cc(G) is symmetric with uU≥0, ∫GuU dmG=1, ∥uU∥1=1, and f∗uU→f in A for every f∈A (Translation continuity and normalised local approximate identities on an LCA group, The axiom of dependent choice: a relation in which every element is related to something admits an N-indexed chain).

[F4]

In A+=C⊕A the spectrum of (0,a) is {0}∪a^(G^) and rA+(0,a)=lim⁡m∥am∥11/m (Scalar unitisation of L^1 of an LCA group: characters, spectrum and identity criterion, Spectral radius formula, Spectral radius).

[F5]

Cc(G) is dense in A every transform lies in C0(G^) by Riemann–Lebesgue, and the transform core is a self-adjoint algebra: products and complex conjugates of transforms of A are transforms of A (C_c(X) is dense in L^p(mu) for a Radon measure, Riemann-Lebesgue lemma on LCA groups, Fourier transform intertwines translation, modulation and convolution); polynomials without constant term approximate the square-root function uniformly on a compact interval (Polynomials are uniformly dense in C([a,b],R) for every closed interval).

[F6]

Tonelli and Fubini apply to the σ-finite products of σ-compact essential supports, and Haar measure is positive on nonempty open sets and finite on compact sets (Tonelli's theorem for nonnegative measurable functions on a sigma-finite product, Fubini's theorem for L^1 functions on a sigma-finite product, Haar measure is positive on nonempty open sets and finite on compact sets, Left Haar integral and left Haar measure).

Proof

technique · direct
1.1F1

(Boundedness.) By [F1], ∣ϕ(−x)∣=∣ϕ(x)∣≤k for every x, so the integral defining Lϕ(f) converges absolutely for every f∈A and ∣Lϕ(f)∣≤k∥f∥1; linearity in f is immediate.

2.1F1F2F5F6

(Integrated positivity.) For f∈A, (f∗f∗)(u)=∫Gf(y)f(y−u)‾ dmG(y), so by inversion invariance of mG the substitution z=y−u in the inner integral (a reflection followed by a translation) gives Lϕ(f∗f∗)=∫G∫Gf(y)f(z)‾ϕ(z−y) dmG(z) dmG(y), and the integrand is carried by the σ-finite product of a σ-compact essential support of f with itself ([F6]). For f∈Cc(G) put K=supp⁡f. The continuous kernel f(y)f(z)‾ϕ(z−y) on K×K admits finite Borel partitions of K on whose product cells its oscillation is arbitrarily small: use compactness and continuity in the group uniformity to take a finite sufficiently small cover, then disjointify it. Choose a point yj in each nonempty cell Pj. The resulting sums ∑j,kcjck‾ϕ(yk−yj), where cj=f(yj)mG(Pj), converge to the double integral since their error is bounded by the kernel oscillation times mG(K)2. Each sum is nonnegative by positive definiteness applied to the points −yj and coefficients cj. Hence Lϕ(f∗f∗)≥0 for f∈Cc(G); for general f∈A choose fn∈Cc(G) with fn→f in A ([F5]), then fn∗fn∗→f∗f∗ in A by [F2] and Lϕ(fn∗fn∗)→Lϕ(f∗f∗) by step 1.1, so the inequality passes to the limit.

3.1F1F2F3step 1.1step 2.1

(The Cauchy-Schwarz bound.) The form [g,h]:=Lϕ(g∗h∗) is sesquilinear by [F2] and positive semidefinite by step 2.1; for such a form ∣[g,h]∣2≤[g,g][h,h] (the quadratic [g+th,g+th]≥0 in t∈C has nonnegative discriminant). With h=uU∗=uU from [F3] this gives ∣Lϕ(f∗uU)∣2=∣[f,uU]∣2≤Lϕ(f∗f∗) Lϕ(uU∗uU∗). As U shrinks, f∗uU→f in A ([F3]) so Lϕ(f∗uU)→Lϕ(f) by step 1.1; and Lϕ(uU∗uU∗)→k: the functions uU∗uU∗ are nonnegative with integral 1 and support shrinking to {0}, so ∣Lϕ(uU∗uU∗)−k∣=∣∫G(uU∗uU∗)(x)(ϕ(−x)−k) dmG(x)∣≤sup⁡x∈supp⁡(uU∗uU∗)∣ϕ(−x)−k∣→0 by continuity of ϕ at 0 and [F1]. Hence ∣Lϕ(f)∣2≤k Lϕ(f∗f∗).

4.1F2F4step 3.1

(The sup-norm bound.) If k=0, [F1] makes Lϕ=0 and all bounds hold. Assume k>0 and put a:=f∗f∗, so a∗=a and a^=∣f^∣2 by [F2]. Applying step 3.1 to a,a2,a4,… gives by induction ∣Lϕ(f)∣≤k1−2−nLϕ(a2n−1)2−n for n≥1, and the elementary bound of step 1.1 applied to the last factor yields ∣Lϕ(f)∣≤k ∥a2n−1∥12−n. Since ∥a2n−1∥12−n=(∥a2n−1∥11/2n−1)1/2→rA+(0,a)1/2 by the spectral radius formula [F4], while rA+(0,a)=max⁡λ∈σA+(0,a)∣λ∣=sup⁡γ∣a^(γ)∣=∥f^∥∞2 by [F4], we obtain ∣Lϕ(f)∣≤k∥f^∥∞. Applying this bound to a=f∗f∗ gives 0≤Lϕ(a)≤k∥a^∥∞=k∥f^∥∞2, proving the second inequality in the stated chain. The bound holds for every representative, so Lϕ vanishes on {f:f^=0}.

5.1F2F5step 2.1step 4.1

(Descent and positivity.) Since Lϕ vanishes on the kernel of the transform, Fϕ(f^):=Lϕ(f) is a well-defined linear functional on the transform core, and step 4.1 gives ∣Fϕ(h)∣≤k∥h∥∞. For positivity let h≥0 belong to the core. The core is a self-adjoint algebra of functions vanishing at infinity ([F5]), so choose real polynomials pn with pn(0)=0 and pn(t)2→t uniformly on [0,∥h∥∞] ([F5]); then pn(h) lies in the core with ∣pn(h)∣2=pn(h)2→h uniformly, and writing pn(h)=gn^ for some gn∈A we get ∣pn(h)∣2=gn∗gn∗^, so Fϕ(∣pn(h)∣2)=Lϕ(gn∗gn∗)≥0 by step 2.1; passing to the uniform limit using the bound of step 4.1 gives Fϕ(h)≥0.

6.1step 1.1step 2.1step 3.1step 4.1step 5.1∎

Steps 1.1, 2.1, 3.1, 4.1 and 5.1 establish every displayed claim: well-definedness and the L1 bound, integrated positivity, the Cauchy-Schwarz bound ∣Lϕ(f)∣2≤kLϕ(f∗f∗)≤k2∥f^∥∞2, the vanishing on {f:f^=0} and the descent to a positive functional with ∣Fϕ(h)∣≤k∥h∥∞.

Depends on

Used by

Dependency tree · two levels

100 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