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.

Local Lp compactness of Wloc1,p-bounded sequences

Statement

Assume the Axiom of Choice. Let n≥1, let Ω⊆Rn be open and let 1≤p<∞. Let (uj) be a sequence that is bounded in Wloc1,p(Ω): for every Ω′⋐Ω there is CΩ′ with ∥uj∥W1,p(Ω′)≤CΩ′ for all j. Then (uj) has a subsequence converging in Llocp(Ω).

If additionally 1<p<∞, the limit of that subsequence can be chosen in Wloc1,p(Ω), and then it is the limit in Llocp(Ω). For p=1 membership of the limit in Wloc1,1(Ω) is not asserted: the weak-compactness argument below uses the reflexivity of Lp, which fails at p=1, and a weak limit of L1 gradients need not be an L1 function. The compactness conclusion itself is proved for every 1≤p<∞.

Facts & Assumptions

Given: the Axiom of Choice, an open set Ω⊆Rn, 1≤p<∞, and a sequence (uj) bounded in Wloc1,p(Ω).

[F1]

Countable relatively compact ball cover. The rational balls B(q,r) with q∈Qn, r∈Q>0 and B(q,r)‾⊆Ω form a countable cover of Ω: given x∈Ω, openness and density of Qn and Q give such a ball containing x. Countability follows from countability of Q and finite products, and each closed ball is compact by Heine--Borel. (The rationals embed densely in the reals, Q is countably infinite, A product of two at most countable sets is at most countable, Heine-Borel in Rn: with the Euclidean metric a subset of Rn is compact if and only if it is closed and bounded, and the proof by bisection uses no choice principle; the same holds on the real line)

[F2]

Rellich on smooth balls. Every bounded ball is a bounded smooth extension domain, and every sequence bounded in W1,p on such a ball has a subsequence converging in Lp. (Bounded C^k domains admit integer-order Sobolev extension, Compactness of W1,p(Ω)↪Lp(Ω) on bounded extension domains)

[F3]

Reflexivity of Lp. Assume Countable Choice. For 1<p<∞ and every measure space, Lp is reflexive, so every norm-bounded sequence in Lp has a weakly convergent subsequence under the ultrafilter lemma, Dependent Choice and Hahn--Banach, all supplied by the Axiom of Choice. (Reflexivity of Lp for one less p less infinity, Reflexivity is equivalent to weak subsequential compactness of bounded sequences, Weak convergence of nets and sequences, The Axiom of Countable Choice (ACω), The axiom of dependent choice: a relation in which every element is related to something admits an N-indexed chain)

[F4]

Weak derivatives pass to weak limits. If uk⇀u and Diuk⇀vi in Lp(B), 1<p<∞, then u∈W1,p(B) with Diu=vi: test against φ∈Cc∞(B) and pass to the limit in ∫uk∂iφ=−∫Diukφ. (Integer-order Sobolev spaces and their norms, Weak derivative of a locally integrable function, Weak convergence of nets and sequences)

Proof

technique · Extract successively on a countable cover by relatively compact rational balls and diagonalise. For $p>1$, use weak compactness on each ball to identify the local limit's weak derivatives
1.1F1F2given

If Ω=∅, the assertions are immediate. Otherwise enumerate the countable cover in [F1] as (Bm)m≥1. The local boundedness hypothesis makes (uj) bounded in W1,p(B1), so [F2] gives a subsequence converging in Lp(B1). Recursively, after obtaining a subsequence converging on B1,…,Bm, apply [F2] to that subsequence on Bm+1 and retain a further subsequence converging there. Countable and Dependent Choice select these nested subsequences. The diagonal sequence (vj), taking the j-th term of the j-th subsequence, is eventually a subsequence of each stage; hence it converges in Lp(Bm) for every m.

2.1F1F5step 1.1

Let um be the Lp(Bm) limit of (vj). On every overlap Bm∩Bℓ the limits um and uℓ agree almost everywhere, by uniqueness of limits of the same sequence in Lp(Bm∩Bℓ). Since the cover is countable, these compatible classes patch to a class u∈Llocp(Ω). If Ω′⋐Ω, compactness and [F5] give a finite subcover Ω′⊆⋃m∈JBm; therefore ∥vj−u∥Lp(Ω′)p≤∑m∈J∥vj−u∥Lp(Bm)p⟶0. Thus vj→u in Llocp(Ω).

3.1F2F3F4step 1.1step 2.1∎

Suppose 1<p<∞ and fix m. The sequence (vj) is bounded in W1,p(Bm). By [F3], after finitely many further subsequence extractions, its function and each of its n weak derivatives converge weakly in Lp(Bm), say vjk⇀w and Divjk⇀wi. Step 2.1 gives strong convergence to um in Lp(Bm), so the weak limit is w=um. Passing to the limit in the weak-derivative identities against each φ∈Cc∞(Bm) and using [F4] gives um∈W1,p(Bm) with Dium=wi. This argument may use a further subsequence depending on m: it identifies the already fixed strong limit um, so it identifies the derivatives of the already fixed limit on every ball without changing the diagonal sequence of step 1.1. On overlaps these derivative classes agree by the weak test identity. For any Ω′⋐Ω, choose a finite ball subcover of Ω′‾ and an ambient smooth partition equal to one near that compact set (Finite ambient partitions near compact sets). Testing after multiplication by the partition pieces proves the weak derivative identity on Ω′; the partition-gradient terms sum to zero, and the finitely many local Lp bounds give global Lp(Ω′) bounds. Thus u∈Wloc1,p(Ω). For p=1 this weak-compactness step is unavailable, and no membership of the limit in Wloc1,1 is asserted. The assumed Axiom of Choice supplies the countable selections in step 1.1 and the Countable and Dependent Choice interfaces of [F2] and [F3].

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

111 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