Alphabeta Math
TheoremStatement: 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.

The Fr'echet--Kolmogorov compactness criterion in Lp(Rn)

Statement

Assume the Axiom of Countable Choice and the Axiom of Dependent Choice. Let 1≤p<∞ and let F⊆Lp(Rn). In each of the three displayed nonnegative suprema, take the value 0 if F=∅. Assume F satisfies: (i) sup⁡f∈F∥f∥Lp(Rn)<∞; (ii) for every ε>0 there is R>0 with sup⁡f∈F(∫∣x∣>R∣f∣p)1/p<ε; and (iii) for every ε>0 there is δ>0 with sup⁡f∈F∥τhf−f∥Lp(Rn)<ε whenever ∣h∣<δ. Then F is totally bounded in Lp(Rn), the closure of F is compact, and every sequence in F has a subsequence converging in Lp(Rn).

Facts & Assumptions

Given: the Axioms of Countable and Dependent Choice, 1≤p<∞, and a family F⊆Lp(Rn) satisfying conditions (i)--(iii) of the statement; write M:=sup⁡f∈F∥f∥p<∞. For R>0 and f∈F put fR:=f1{∣x∣≤R}, and for each ε>0 let ηε be the radial mollifier at scale ε of A radial mollifier family in Rn.

[F1]

Mollification. For fR∈Lp⊆Lloc1, the function fR∗ηδ is smooth and ∂α(fR∗ηδ)=fR∗(∂αηδ). (Convolution with a mollifier is smooth, and derivatives pass under the integral sign)

[F2]

H"older's inequality. For conjugate exponents p,p′, ∫∣uv∣≤∥u∥p∥v∥p′; in particular ∣(fR∗ηδ)(x)∣≤∥fR∥p∥ηδ∥p′ and ∣∇(fR∗ηδ)(x)∣≤∥fR∥p∥∇ηδ∥p′ pointwise. (Holder's inequality for integrals, including the endpoint cases)

[F3]

Minkowski's integral inequality. For measurable F on a product of sigma-finite measure spaces with ∫Y∥F(⋅,y)∥p dν(y)<∞, ∥∫YF(⋅,y) dν(y)∥p≤∫Y∥F(⋅,y)∥p dν(y). (Minkowski's integral inequality)

[F4]

Translations are Lp-isometries. ∥τhg∥p=∥g∥p for every g∈Lp and every h, since Lebesgue measure is translation invariant. (Translation of a function on Rn, Lebesgue outer measure, Lebesgue measurability and Lebesgue measure are unchanged by translation, The space Lp(μ) as the quotient by null functions)

[F5]

Finite nets for equicontinuous families. An equicontinuous pointwise bounded family in C(K,R), K a compact metric space, is totally bounded for the supremum metric; the same holds after applying the statement to real and imaginary parts of a complex-valued family, and Arzelà--Ascoli for real C(K) under Countable Choice and Dependent Choice: compact closure iff equicontinuous and pointwise bounded records the equivalent compact-closure form. (An equicontinuous pointwise-bounded family in C(K,R) has a finite net in the supremum metric, Arzelà--Ascoli for real C(K) under Countable Choice and Dependent Choice: compact closure iff equicontinuous and pointwise bounded)

[F6]

Total boundedness and its closure. A metric space is totally bounded when it has a finite δ-net for every δ>0; total boundedness passes to subsets and to closures, and for g,h supported in a compact ball K one has ∥g−h∥p≤∥g−h∥∞∣K∣1/p. (Finite ε-net and totally bounded metric space, A totally bounded metric space is bounded, every subspace of a totally bounded space is totally bounded, and the closure of a totally bounded subset is totally bounded, Open ball, closed ball and sphere in a metric space)

Proof

technique · Make the tails small and the translations small, regularise by convolution, use Arzel\`a--Ascoli to get a finite net for the regularised family, and transfer the net back to $\mathcal F$
1.1given

If F=∅ it is totally bounded and its closure is empty. Otherwise fix ε>0. By (ii) choose R>0 with ∥f−fR∥p<ε for every f∈F, and by (iii) choose δ>0 with ∥τhf−f∥p<ε for every f∈F and every ∣h∣<δ; let η:=ηδ/2 be the radial mollifier at scale δ/2, so ∫η=1, η≥0 and supp⁡η⊆B(0,δ/2).

2.1F3F4step 1.1

For every f∈F and every ∣y∣<δ, ∥fR−τyfR∥p≤∥fR−f∥p+∥f−τyf∥p+∥τyf−τyfR∥p<3ε by step 1.1 and [F4]. Since fR−fR∗η=∫η(y)(fR−τyfR)dy, [F3] gives ∥fR−fR∗η∥p≤∫η(y)∥fR−τyfR∥p dy<3ε, and therefore ∥f−fR∗η∥p<4ε.

2.2F1F2step 1.1

By [F1] and [F2], every g=fR∗η satisfies ∥g∥∞≤M∥η∥p′ and ∥∇g∥∞≤M∥∇η∥p′, and g vanishes off the ball of radius R+δ because fR vanishes off the ball of radius R; hence the family G:={fR∗η:f∈F} is uniformly bounded and uniformly Lipschitz, and all its elements are supported in the compact ball K:=B(0,R+δ)‾.

3.1F5F6step 2.2

The real parts {Re⁡g:g∈G} are equicontinuous and pointwise bounded on the compact metric space K, and likewise the imaginary parts; by [F5] both families are totally bounded in the supremum metric, and combining the finitely many real and imaginary sup-balls, G is totally bounded in the supremum metric over K. Since every g is supported in K, the supremum over Rn equals the supremum over K, so for every ϑ>0 the family G has a finite covering by Lp-balls of radius ϑ by [F6].

4.1F6step 2.1step 3.1

Let r>0 and apply steps 1.1--3.1 with ε<r/10, covering G by finitely many Lp-balls of radius r/10. Step 2.1 then covers F by the same centres with radius r/2. Discard empty intersections with F and choose one point of F in each remaining ball. Their radius-r balls cover F by the triangle inequality, so the centres belong to F as required by [F6]. Thus F is totally bounded.

5.1F6F7step 1.1step 4.1∎

By [F6] the closure F‾ is totally bounded, and it is closed in the complete space Lp(Rn), hence complete by [F7]; a complete and totally bounded metric space is compact by [F7], and compactness is equivalent to sequential compactness by [F7], so every sequence in F has a subsequence converging in Lp(Rn). The empty case was disposed of in step 1.1, and no other choice principle is used: Countable Choice covers the completeness-to-compactness step and the finite-net selection, Dependent Choice covers the compactness-sequential equivalence.

Depends on

Used by

Dependency tree · two levels

107 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