Errata (x), pp. 28–29 (claim of p. 19) — β(µ)((v, w), s) = 0 for ρ-a.e. w, all v ∈ ℝᴷ and all s
ProvedDistCov.Hilbert.errata_x_continuityLet be Borel probability measures on with finite first moments and let
be the barycenter of under the Gaussian Crofton embedding. Fix . If holds -almost everywhere, then for -a.e. ,
where is the sequence with first coordinates and remaining coordinates those of .
This is the claim at the end of the penultimate paragraph of the proof of Theorem 3.16, whose justification is supplied in Errata (x): it upgrades "for -a.e. and -a.e. " to "for all and all ".
Formalization Note. Indices start at , so takes coordinates from and coordinates from . The paper pairs the tail of with 's tail coordinates; under the IID law either pairing has the same distribution. The statement starts from the joint a.e. hypothesis on , as the paper does; the passage to almost every and (Fubini, using the product structure of ) is part of what is to be proved. is allowed; the paper has .
import Mathlib import Definitions.Def_DistCov_Hilbert_Setting
namespace DistCov.Hilbert
open MeasureTheory ProbabilityTheory
theorem errata_x_continuity (μ₁ μ₂ : Measure DistCov.Indep.L2N) [IsProbabilityMeasure μ₁]
[IsProbabilityMeasure μ₂] (h₁ : DistCov.Indep.FiniteFirstMoment μ₁) (h₂ : DistCov.Indep.FiniteFirstMoment μ₂) (K : ℕ)
(h : baryDiff μ₁ μ₂ =ᵐ[rho.prod volume] 0) :
∀ᵐ w ∂rho, ∀ (v : Fin K → ℝ) (s : ℝ), baryDiff μ₁ μ₂ (splice K v w, s) = 0 := by sorry
end DistCov.Hilbert
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.