§5.12, Theorem 1 — Mazur geometric Hahn–Banach theorem
DisprovedVectorSpaceOpt.mazur_geometric_hahn_banachLet be a real normed space, a convex set with nonempty interior, and an affine subspace that contains no point of interior . Then there exist a nonzero continuous linear functional and a scalar such that
Thus one affine hyperplane contains all of V while the interior of K lies strictly on one side. This is the chapter's geometric Hahn–Banach form and serves as the common separation engine behind the supporting-hyperplane and two-set separation milestones.
import Mathlib
namespace VectorSpaceOpt
theorem mazur_geometric_hahn_banach
{X : Type*} [NormedAddCommGroup X] [NormedSpace ℝ X]
(K : Set X) (hK : Convex ℝ K) (hKint : (interior K).Nonempty)
(V : AffineSubspace ℝ X) (hV : ∀ v ∈ V, v ∉ interior K) :
∃ (f : X →L[ℝ] ℝ) (c : ℝ),
f ≠ 0 ∧ (∀ v ∈ V, f v = c) ∧ (∀ k ∈ interior K, f k < c) := by sorry
end VectorSpaceOptRead-back
What the Lean code literally says, in plain math · gpt-5
Let be any real normed vector space, let be convex with nonempty interior, and let be a real affine subspace of such that every lies outside the interior of . Then there exist a nonzero continuous real-linear functional and a real number such that for every and for every in the interior of . The hypothesis does not say that lies in or on its boundary, and the conclusion makes no inequality claim for points of outside its interior; no closedness, completeness, finite-dimensionality, or nontriviality assumption is present.
Confirmed by the mission captain (proposal self-audit).