Existence and uniqueness of the Reeb vector in a finite-dimensional contact vector space
ProvedContactLinearAlgebra.reeb_exists_uniquecontact-geometrylinear-algebrasymplectic-geometry
Let V be a finite-dimensional real vector space, a a nonzero linear functional, and b an alternating bilinear form. Assume the restriction of b to ker a is non-degenerate: if u lies in ker a and b(u,v)=0 for every v in ker a, then u=0. There is a unique vector R satisfying
This is the pointwise linear-algebra form of the Reeb-vector theorem, usable independently of any manifold or regular-level representation.
Preamble
import Mathlib.LinearAlgebra.Dual.Lemmas import Mathlib.Data.Real.Basic
Formal statement
theorem ContactLinearAlgebra.reeb_exists_unique {V : Type*} [AddCommGroup V] [Module ℝ V]
[FiniteDimensional ℝ V] (a : V →ₗ[ℝ] ℝ) (b : V →ₗ[ℝ] V →ₗ[ℝ] ℝ)
(hb : ∀ v, b v v = 0) (ha : ∃ v, a v ≠ 0)
(hn : ∀ u, a u = 0 → (∀ v, a v = 0 → b u v = 0) → u = 0) :
∃! R : V, a R = 1 ∧ ∀ v, b R v = 0 := by sorry
Source
Geiges, Contact geometry, https://arxiv.org/pdf/math/0307242, Remark 2.3 (p. 4) and Definition 2.5 with following paragraph (p. 6). Abstract pointwise linear-algebra formulation using non-degeneracy on the contact hyperplane; no manifold hypotheses are required.