The neighborhood criterion permits finitely many exceptional measures
ProvedQAlgorithms.Papers.Independent.Support50.P33.tightnessNeighborhoodFiniteExceptionsLet be a real Hausdorff locally convex topological vector space, and let be convex and von Neumann bounded. Assume the relative topology of is generated by a countable subfamily of the ambient defining seminorms, and that is complete and separable in the resulting metrizable additive uniformity. The set is allowed to be empty. All regularity and convergence below are relative to .
Write for the Borel probability measures on with the weak convergence topology. The barycenter is the Pettis integral of inclusion: every continuous real linear functional is integrable against , and . The set is -compact when the barycenter preimage of every compact subset of is compact in ; it is pointwise -compact when every barycenter fiber is compact.
Fix a complete metric inducing the inherited additive uniformity on , and let be the closed -neighborhood, with . For an arbitrary subset ,
Both and the finite exceptional set may depend on and , and both are chosen before the universal measure. Here tightness has its usual uniform compact-complement meaning. No convexity or barycenter constraint is imposed on .
This form allows the neighborhood test to ignore finitely many measures for each pair of parameters.
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P33
noncomputable section
open Set Filter MeasureTheory TopologicalSpace
open scoped Topology ENNReal
namespace QAlgorithms.Papers.Independent.Support50.P33
open QAlgorithms.Papers.Independent.Frontier18.P10
open QAlgorithms.Papers.Independent.Dependency50.P25
attribute [local instance] QAlgorithms.Papers.Independent.Support50.P33.paperMeasurable
attribute [local instance] QAlgorithms.Papers.Independent.Support50.P33.paperBorel
section LocallyConvexResults
variable {E : Type*} [UniformSpace E] [AddCommGroup E] [Module ℝ E]
[hgroup : IsUniformAddGroup E] [hsmul : ContinuousSMul ℝ E]
[hlc : LocallyConvexSpace ℝ E] [hhaus : T2Space E]
include hgroup hsmul hlc hhaus
/-- Unnumbered consequence after Lemma 2, arXiv:1004.3792v1, PDF p.5.
The finite exceptional subset of M, as well as the compact witness, may
vary with epsilon and delta; both precede the universal measure binder. -/
theorem tightnessNeighborhoodFiniteExceptions (A : Set E)
(hA : GeneralBarycenterSpace A) (hc : Convex ℝ A)
(d : MetricSpace A) (hd : d.toUniformSpace = (inferInstance : UniformSpace A))
(M : Set (ProbabilityMeasure A)) :
IsTightMeasureSet ((fun mu : ProbabilityMeasure A => (mu : Measure A)) '' M) ↔
∀ epsilon : ℝ, 0 < epsilon → ∀ delta : ℝ, 0 < delta →
∃ K : Set A, IsCompact K ∧
∃ F : Set (ProbabilityMeasure A), F ⊆ M ∧ F.Finite ∧
∀ mu ∈ M \ F, ENNReal.ofReal (1 - epsilon) ≤
(mu : Measure A) (letI : MetricSpace A := d; Metric.cthickening delta K) := by sorry
end LocallyConvexResults
end QAlgorithms.Papers.Independent.Support50.P33
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be the ambient real vector space with its uniform structure and induced topology, and let be closed, convex, and von Neumann bounded, with complete, separable, and metrizable. Let be any metric on whose induced uniform structure equals the given subtype uniform structure, and let be any set of Borel probability measures on . The underlying measures in are uniformly tight—meaning that, for every real , there is a compact with for every —if and only if, for every real and every real , there exist a compact and a finite set such that for every . Here is the closed -neighborhood computed with , consisting of points at distance at most from , and is empty if is empty; the inequality is in the extended nonnegative reals. Both and may depend on and , and may be empty or may equal when is finite, making the remaining universal condition vacuous. Empty and empty are also permitted, and the lower bound is zero whenever .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.