Alphabeta Math
TheoremStatement: AI-adaptedProof: AI-adaptedPipeline-generatedprecheck passjudge pass (gpt-5.6-terra)audited 2026-09-09
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.

Relative norm-preserving Hahn–Banach extension over the real and complex fields

Statement

Assume HB. Let X be a normed space over K∈{R,C}, M≤X any K-linear subspace, and g:M→K a bounded K-linear functional. There exists F∈X∗ such that F∣M=g and ∥F∥=∥g∥. The subspace need not be closed, and X need not be complete; M={0} is allowed.

Facts & Assumptions

[F1]

Under HB a real dominated functional extends with the two signed bounds (Dominated extension conditional on the relative principle).

[F2]

The dual consists of bounded scalar-linear functionals, with norm sup⁡∥x∥≤1∣f(x)∣ (The dual space X^* of a normed space and its dual norm).

[F3]

A real-linear u reconstructs a complex-linear f(x)=u(x)−iu(ix) with real part u, and reconstructs any complex-linear functional from its real part (A complex linear functional is recovered from its real part by f(x)=u(x)-iu(ix)).

[F4]

A linear subspace contains zero and is closed under addition and scalar multiplication (Linear subspace of a vector space).

Proof

Given: HB, a normed K-space X, a K-linear subspace M, and bounded g:M→K.

1.1givenF2F4algebra

Put C=∥g∥≥0. For m≠0, the vector m/∥m∥ is in the unit ball of M, so ∣g(m)∣=∥m∥∣g(m/∥m∥)∣≤C∥m∥; for m=0 the same inequality holds because g(0)=0. Set p(x)=C∥x∥. Then p(tx)=tp(x) for t≥0 and p(x+y)≤p(x)+p(y) by the norm axioms.

2.1step 1.1F1F2

Over R, g≤p∣M by the preceding estimate. The real extension theorem gives F∣M=g and −C∥x∥≤F(x)≤C∥x∥. Thus ∣F(x)∣≤C∥x∥, so F∈X∗ and ∥F∥≤C.

2.2step 1.1F1F3F4

Over C, the underlying real space of M is a real linear subspace of the underlying real X, because closure under complex scalars includes closure under real scalars. Let u=Re⁡g. It is real linear and u(m)≤∣g(m)∣≤p(m). The real extension theorem gives real-linear U:X→R with U∣M=u and U≤p.

3.1step 2.2F3F4

Define F(x)=U(x)−iU(ix). The reconstruction lemma gives complex linearity and Re⁡F=U. For m∈M, also im∈M, whence F(m)=u(m)−iu(im)=g(m) by the same lemma applied to g.

4.1step 2.2step 3.1F2algebra

If F(x)=0 then ∣F(x)∣≤C∥x∥. Otherwise set a=F(x)‾/∣F(x)∣. Then ∣a∣=1 and F(ax)=aF(x)=∣F(x)∣ is real, so ∣F(x)∣=U(ax)≤C∥ax∥=C∥x∥. Consequently the complex extension is bounded and ∥F∥≤C.

5.1step 2.1step 3.1step 4.1F2∎

In either field, F extends g. For every m∈M with ∥m∥≤1, ∣g(m)∣=∣F(m)∣≤∥F∥; taking the supremum gives C≤∥F∥. Together with the upper bounds this yields ∥F∥=∥g∥. If C=0, the bound forces F=0; in particular this covers M={0} and the zero space.

Source notes

Brezis Corollary 1.2, p.3 (real); Teschl Theorem 4.14 and Corollary 4.15, pp.113–114.

Depends on

Used by

Dependency tree · two levels

16 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