Alphabeta Math
TheoremStatement: Literature-sourcedProof: Literature-sourcedPipeline-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 half-space trace estimate and the half-space trace operator

Statement

Assume the Axiom of Choice. Let n≥2, d=n−1, 1≤p<∞, K∈{R,C}, and let H={x=(x′,xn)∈Rn:xn>0} with ∂H=Rd×{0}. Write ∂n for the last coordinate derivative.

(i) For every u∈C(H‾)∩W1,p(H;K) with compact support, the classical boundary function g=u(⋅,0) satisfies ∣g(x′)∣p≤p∫0∞∣u(x′,t)∣p−1 ∣∂nu(x′,t)∣ dt for a.e. x′∈Rd, hence ∥g∥Lp(Rd)p≤p ∥u∥Lp(H)p−1∥∂nu∥Lp(H)≤p ∥u∥W1,p(H)p. For every h>0 the strip form h ∣g(x′)∣p≤Cp(∫0h∣u(x′,t)∣pdt+hp∫0h∣∂nu(x′,t)∣pdt) holds for a.e. x′, with Cp=2p−1 for 1<p<∞ and C1=1.

(ii) There is a unique bounded linear operator T+:W1,p(H;K)→Lp(Rd;K) with T+u=u(⋅,0) for every compactly supported u∈C(H‾)∩W1,p(H), and ∥T+u∥Lp(Rd)≤C(n,p)∥u∥W1,p(H); it is the unique bounded extension of classical restriction, and T+u depends only on the a.e. class of u.

Facts & Assumptions

Given: The Axiom of Choice; n≥2, d=n−1, 1≤p<∞; the half-space H and its boundary; the space W1,p(H;K) with norm ∥u∥W1,p(H) of Integer-order Sobolev spaces and their norms; and the class convention of The space Lp(μ) as the quotient by null functions.

[F1]

Assume the Axiom of Choice through its Countable-Choice and Dependent-Choice interfaces. For open Ω⊆Rn, n≥1, and 1≤p<∞: u∈W1,p(Ω;K) if and only if u∈Lp(Ω;K) has one measurable ACL representative u∗ whose classical coordinate derivatives exist almost everywhere, are measurable and lie in Lp; then ∂iu∗ represents Diu almost everywhere. (The ACL characterisation of W1,p, The Axiom of Choice)

[F2]

Assume the Axiom of Choice through the ACL interface. For a bounded interval I=(a,b), 1≤p<∞ and u∈W1,p(I;K) with weak derivative u′, the absolutely continuous representative u∗ satisfies, for every 0<ε<b−a, the endpoint inequality, in particular ε∣u∗(a)∣p≤2p−1(∫aa+ε∣u∗∣p+εp∫aa+ε∣u′∣p) for 1<p<∞ and ε∣u∗(a)∣≤∫aa+ε∣u∗∣+ε∫aa+ε∣u′∣ for p=1. (The one-dimensional endpoint estimate on a bounded interval)

[F3]

Holder's inequality: for conjugate exponents p,p′ and measurable φ,ψ with φ∈Lp, ψ∈Lp′, ∫∣φψ∣≤∥φ∥p∥ψ∥p′, so φψ is integrable. (Holder's inequality for integrals, including the endpoint cases)

[F4]

Assume Countable Choice. For a nonnegative measurable function on a product of sigma-finite measure spaces the double integral equals the two iterated integrals, with measurable section integrals. (Tonelli and Fubini for the completed product, with only almost-everywhere section measurability)

[F5]

There is a bounded linear extension operator E:W1,p(H;K)→W1,p(Rn;K) with (Eu)∣H=u almost everywhere on H. (Integer-order Sobolev extension from a half-space)

[F6]

Assume Countable Choice. Cc∞(Rn;K) is dense in W1,p(Rn;K). (Compactly supported smooth functions are dense in W^{k,p}(R^n))

[F7]

Assume Countable Choice. Lp(Rd;K) is complete, and a norm-convergent sequence has an almost-everywhere convergent subsequence. (Riesz-Fischer completeness of Lp for 1≤p≤∞)

[F8]

W1,p(H;K) is a complete normed space. (Integer-order Sobolev spaces are Banach)

[F9]

Assume Countable Choice. Let X be a normed space with completion (X^,i) and let Y be a Banach space; a linear T:X→Y with ∥Tx∥≤C∥x∥ extends uniquely to a bounded linear T^:X^→Y with ∥T^u∥≤C∥u∥. (Bounded linear maps extend uniquely across the completion)

Proof

technique · direct
1.1F1algebragiven

The pointwise normal-line identity. Fix a compactly supported u∈C(H‾)∩W1,p(H). By [F1] applied to Ω=H there is an ACL representative u∗ of the class of u whose classical last derivative ∂tu∗ exists a.e., lies in Lp(H) and represents ∂nu. For a.e. x′∈Rd the section t↦u∗(x′,t) is absolutely continuous on compact subintervals of (0,∞), and u∗=u a.e. on H; since both the continuous extension of the section (which exists because ∫0T∣∂tu∗∣dt<∞ for every T) and the continuous function u(x′,⋅) agree on a dense set of t, they agree everywhere on the line, so the section extends continuously to t=0 with value g(x′)=u(x′,0). Because u has compact support, the section vanishes for large t, and the absolutely continuous function t↦∣u∗(x′,t)∣p satisfies ∣g(x′)∣p=−∫0∞∂t∣u∗∣pdt=−p∫0∞∣u∗∣p−2Re⁡(u∗‾∂tu∗)dt≤p∫0∞∣u∣p−1∣∂nu∣dt for 1<p<∞, while for p=1 the absolutely continuous function t↦∣u∗(x′,t)∣ has ∣g(x′)∣=−∫0∞∂t∣u∗∣dt≤∫0∞∣∂nu∣dt.

1.2F5F6algebragiven

Density of the smooth restriction class. Put D:={φ∣H:φ∈Cc∞(Rn;K)}, a linear subspace of W1,p(H), and let u∈W1,p(H) and δ>0. By [F5] there is Eu∈W1,p(Rn) with (Eu)∣H=u a.e.; by [F6] choose φ∈Cc∞(Rn) with ∥φ−Eu∥W1,p(Rn)<δ. Then φ∣H∈D, and because the restriction to H of an Lp class has ∥ψ∣H∥Lp(H)≤∥ψ∥Lp(Rn) componentwise, ∥φ∣H−u∥W1,p(H)≤∥φ−Eu∥W1,p(Rn)<δ. Hence D is dense in W1,p(H).

2.1F1F2F3F4step 1.1algebra

The integrated estimate and the strip form. Integrate the pointwise inequality of step 1.1 over x′∈Rd and use Tonelli [F4] to interchange the x′- and t-integrals: ∥g∥pp≤p∫H∣u∣p−1∣∂nu∣. For p=1 this gives ∥g∥1≤∥∂nu∥1≤∥u∥W1,1; for 1<p<∞, Holder [F3] with exponents p′ and p gives ∫H∣u∣p−1∣∂nu∣≤∥u∥Lpp−1∥∂nu∥Lp≤∥u∥W1,pp, and therefore ∥g∥Lpp≤p∥u∥Lpp−1∥∂nu∥Lp≤p∥u∥W1,pp. For the strip form, fix h>0; for a.e. x′ the section of u∗ on (0,2h) is absolutely continuous with Lp derivative, so [F1] in dimension one makes that section an element of W1,p((0,2h)) with weak derivative ∂nu(x′,⋅), and [F2] applies with ε=h and gives, after multiplying by h, h∣g(x′)∣p≤Cp(∫0h∣u(x′,t)∣pdt+hp∫0h∣∂nu(x′,t)∣pdt) with the stated Cp.

3.1F7F8F9step 1.2step 2.1algebra

Construction of T+ and agreement with classical restriction. Let S:D→Lp(Rd) be the classical restriction Sφ:=φ(⋅,0), which is linear and, by step 2.1 applied to φ∈D⊂C(H‾)∩W1,p(H) (each is continuous on H‾ with compact support), satisfies ∥Sφ∥p≤p1/p∥φ∥W1,p(H). Since D is dense in W1,p(H) by step 1.2 and carries the subspace norm, and since W1,p(H) is complete by [F8], the pair (W1,p(H),inclusion) is a completion of D; by the completion universal property [F9] applied with Y=Lp(Rd) (complete by [F7]), S extends uniquely to a bounded linear T+:W1,p(H)→Lp(Rd) with ∥T+u∥p≤p1/p∥u∥W1,p(H) and T+∣D=S. If now u∈C(H‾)∩W1,p(H) is compactly supported, choose φm∈D with φm→u in W1,p(H); then T+u=lim⁡mSφm in Lp by continuity, while ∥Sφm−u(⋅,0)∥pp≤p∥φm−u∥W1,p(H)p→0 by step 2.1 applied to the compactly supported continuous difference, so T+u=u(⋅,0) in Lp(Rd).

4.1step 1.2step 3.1algebragiven∎

Uniqueness and class-dependence. If T′ is another bounded linear operator on W1,p(H) whose restriction to D is S, then A:=T+−T′ is a bounded linear operator vanishing on D; for u∈W1,p(H) choose φm∈D with φm→u (step 1.2), so ∥Au∥=lim⁡m∥Aφm∥=0 by boundedness of A. Hence T′=T+: this is the asserted uniqueness of the bounded extension of classical restriction. Moreover, if u=v in W1,p(H) are the same a.e. class, then u−v is the zero class and linearity gives T+(u−v)=T+0=0, because the zero class is the limit of the constant sequence 0∈D and S0=0; hence T+u=T+v and T+ depends only on the class. This proves (i) and (ii).

Source notes

Hunter's Theorem 3.44 and its proof (printed pp. 71-73) proves the pointwise normal-line inequality and the bounded half-space trace; Laugesen's flat estimate and dense-subspace extension (Theorem 3.14, printed pp. 62-64), Schikorra's Theorem III.3.21 (printed pp. 76-77) and Teschl's Theorem 9.18 (printed pp. 208-209) are independent treatments of the same construction. The density of the smooth restriction class uses the published half-space extension operator and the interior density theorem on Rn; this replaces the scaffold's route through the bounded domains H∩BR, whose boundaries have corners and so are not covered by the bounded-C1-domain density theorem.

Depends on

Used by

Dependency tree · two levels

88 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