Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

(3.4), p. 11 — D(µ) = −2‖β(µ)‖² when µ(X) = 0

Proved
DistCov.Indep.eq_3_4

by mikedeng1 · Oct 9, 2026 · Mathlib 0df444a (Lean v4.33.1)

barycenterhilbert-spacenegative-typep2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Let X\mathcal XX be a separable metric space with its Borel σ\sigmaσ-field, HHH a real Hilbert space, and ϕ:X→H\phi:\mathcal X\to Hϕ:X→H an embedding, i.e. d(x,x′)=∥ϕ(x)−ϕ(x′)∥2d(x,x')=\|\phi(x)-\phi(x')\|^2d(x,x′)=∥ϕ(x)−ϕ(x′)∥2 for all x,x′x,x'x,x′. Let m1,m2m_1,m_2m1​,m2​ be finite measures on X\mathcal XX with finite first moments and equal total mass, m1(X)=m2(X)m_1(\mathcal X)=m_2(\mathcal X)m1​(X)=m2​(X), and put μ=m1−m2\mu=m_1-m_2μ=m1​−m2​, a signed measure with μ(X)=0\mu(\mathcal X)=0μ(X)=0. Then

D(μ)=−2 ∥βϕ(μ)∥2,i.e.D(m1−m2)=−2 ∥βϕ(m1)−βϕ(m2)∥2,D(\mu)=-2\,\|\beta_\phi(\mu)\|^2,\qquad\text{i.e.}\qquad D(m_1-m_2)=-2\,\big\|\beta_\phi(m_1)-\beta_\phi(m_2)\big\|^2,D(μ)=−2∥βϕ​(μ)∥2,i.e.D(m1​−m2​)=−2​βϕ​(m1​)−βϕ​(m2​)​2,

where βϕ(m)=∫ϕ dm\beta_\phi(m)=\int\phi\,dmβϕ​(m)=∫ϕdm is the barycenter.

Identity (3.4) converts the energy DDD of a mass-zero signed measure into a squared Hilbert norm; it shows at once that negative type gives D(μ1−μ2)≤0D(\mu_1-\mu_2)\le0D(μ1​−μ2​)≤0 and is the key step to Proposition 3.1.

Formalization Note Every finite signed measure μ\muμ with μ(X)=0\mu(\mathcal X)=0μ(X)=0 whose variation has finite first moment is m1−m2m_1-m_2m1​−m2​ for its Jordan parts, which have equal mass and finite first moments, so the statement for pairs is (3.4) itself; D(m1−m2)D(m_1-m_2)D(m1​−m2​) is the bilinear expansion Ddiff, and βϕ(m1−m2)=βϕ(m1)−βϕ(m2)\beta_\phi(m_1-m_2)=\beta_\phi(m_1)-\beta_\phi(m_2)βϕ​(m1​−m2​)=βϕ​(m1​)−βϕ​(m2​) by linearity. The integral defining βϕ\beta_\phiβϕ​ is genuine: ϕ\phiϕ is continuous (∥ϕ(x)−ϕ(x′)∥=d(x,x′)1/2\|\phi(x)-\phi(x')\|=d(x,x')^{1/2}∥ϕ(x)−ϕ(x′)∥=d(x,x′)1/2), hence strongly measurable on the separable space, and ∥ϕ(x)∥≤∥ϕ(o)∥+1+d(o,x)\|\phi(x)\|\le\|\phi(o)\|+1+d(o,x)∥ϕ(x)∥≤∥ϕ(o)∥+1+d(o,x) is integrable. Separability is the standing assumption of Errata (i), p. 24.

Preamble
import Mathlib
import Definitions.Def_DistCov_Indep_Setting
Formal statement
namespace DistCov.Indep

open MeasureTheory

theorem eq_3_4 {X : Type*} [MetricSpace X] [MeasurableSpace X] [BorelSpace X] [SecondCountableTopology X] {H : Type*} [NormedAddCommGroup H] [InnerProductSpace ℝ H] [CompleteSpace H]
    (ϕ : X → H) (hϕ : IsNegTypeEmbedding ϕ)
    (m₁ m₂ : Measure X) [IsFiniteMeasure m₁] [IsFiniteMeasure m₂]
    (h₁ : FiniteFirstMoment m₁) (h₂ : FiniteFirstMoment m₂) (hmass : m₁ Set.univ = m₂ Set.univ) :
    Ddiff m₁ m₂ = -2 * ‖bary ϕ m₁ - bary ϕ m₂‖ ^ 2 := by sorry

end DistCov.Indep
Source
Lyons, Distance covariance in metric spaces, arXiv:1106.5758 (version of 19 Dec. 2020), p. 11, display (3.4) and the sentence before it; Errata (i), p. 24
Human review
  • Endorsed by Shuze Chen · Oct 9, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 9, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me