Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-generatedPipeline-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.

Weakly singular fractional Grönwall inequality

Statement

Let T>0, 0<α<1, A,L≥0, and let y∈C([0,T]) be nonnegative. If y(t)≤A+LIαy(t)(0≤t≤T), then for every t∈[0,T], y(t)≤A∑k=0∞LktkαΓ(kα+1). The k=0 term is 1, including when L=0, and the series converges uniformly on [0,T]. In particular, if A=0 then y≡0.

Facts & Assumptions

Given: T>0, 0<α<1, A,L≥0, y∈C([0,T]), y≥0, and the stated pointwise inequality.

[F1]

Fractional integral and Caputo derivative: for continuous g, Iβg(t)=Γ(β)−1∫0t(t−s)β−1g(s) ds; its kernel is nonnegative for β>0.

[F2]

Semigroup law for fractional integrals: for continuous g, Ia(Ibg)=Ia+bg at every t, and Ic1(t)=tc/Γ(c+1) for a,b,c>0.

[F3]

The real Gamma function by Euler's integral and The real Gamma functional equation Γ(s+1)=sΓ(s): for x>0, Γ(x+1)=∫0∞e−ssx ds>0, and Γ(1)=1.

[F5]

The Cauchy-sequence reals are Archimedean: for every real R there is a natural number k≥1 with R<k.

Proof

1.1givenF3F4F5F6algebra

For x>0, restricting the Euler integral [F3] to [x,x+1] gives Γ(x+1)≥e−(x+1)xx, since the interval has length one and e−ssx≥e−(x+1)xx there. With x=kα and k≥1, this yields LkTkα/Γ(kα+1)≤e(B/kα)k, where B:=LTαeα/αα. If L=0, every term with k≥1 vanishes. Otherwise B>0; apply [F5] to R=(2B)1/α to choose k0>R. For k≥k0, [F4] gives kα≥k0α>(2B)(1/α)α=2B, so (B/kα)k<2−k. The comparison test [F6] gives convergence of the scalar majorant tail. For any p>q≥k0 and any t∈[0,T], the absolute value of the series segment from q to p is at most e∑k=qp2−k. This bound tends to zero independently of t, so the series is uniformly Cauchy on [0,T] and hence converges uniformly there.

1.2givenF1F2induction

The hypothesis is the iteration formula for m=1. If it holds for m≥1, positivity of the kernel [F1] allows applying LmImα to y≤A+LIαy in its remainder, giving LmImαy(t)≤ALmImα1(t)+Lm+1Imα(Iαy)(t)=ALmtmα/Γ(mα+1)+Lm+1I(m+1)αy(t) by [F2]. Substitution proves by induction that y(t)≤A∑k=0m−1Lktkα/Γ(kα+1)+LmImαy(t) for every m≥1.

2.1step 1.1step 1.2F1F2algebra∎

By [F1]–[F2], 0≤LmImαy(t)≤∥y∥∞LmTmα/Γ(mα+1). The last factor tends to zero as the mth term of the convergent series in step 1.1. Letting m→∞ in step 1.2 proves the claimed bound; if A=0, then 0≤y(t)≤0 for every t, so y≡0.

Depends on

Used by

Dependency tree · two levels

65 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