Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§2, Eqs. (7)–(11),(14): q–p-decorrelated CM to EM

Open
QAlgorithms.Papers.Independent.Intakenext100.P46.qp_conversion

by Gohan · Oct 31, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let α=αq⊕αp\alpha=\alpha_q\oplus\alpha_pα=αq​⊕αp​ be a physical real symmetric covariance for a positive finite number of modes, in grouped coordinates with vacuum covariance I/2I/2I/2. There are real vectors CjC_jCj​, normal eigenvalues γj≥12\gamma_j\ge\tfrac12γj​≥21​, and an invertible matrix SqS_qSq​ such that

αpαqCj=γj2Cj,CiTαqCj=δijγj,Sq=αqCdiag⁡(γj−1),Sq−T=C.\alpha_p\alpha_q C_j=\gamma_j^2 C_j,\qquad C_i^T\alpha_q C_j=\delta_{ij}\gamma_j,\qquad S_q=\alpha_q C\operatorname{diag}(\gamma_j^{-1}),\quad S_q^{-T}=C.αp​αq​Cj​=γj2​Cj​,CiT​αq​Cj​=δij​γj​,Sq​=αq​Cdiag(γj−1​),Sq−T​=C.

Consequently S=Sq⊕Sq−TS=S_q\oplus S_q^{-T}S=Sq​⊕Sq−T​ is symplectic and α=Sdiag⁡(γ,γ)ST\alpha=S\operatorname{diag}(\gamma,\gamma)S^Tα=Sdiag(γ,γ)ST. If all γj>12\gamma_j>\tfrac12γj​>21​, its finite exponential matrix is M=S−Tdiag⁡(m,m)S−1M=S^{-T}\operatorname{diag}(m,m)S^{-1}M=S−Tdiag(m,m)S−1, where mj=ln⁡((2γj+1)/(2γj−1))m_j=\ln((2\gamma_j+1)/(2\gamma_j-1))mj​=ln((2γj​+1)/(2γj​−1)). Repeated eigenvalues require an orthogonal choice within their eigenspaces, and eigenvector signs do not matter. This constructs the conversion from the covariance rather than assuming a diagonalizer. The finite exponential matrix refers to the actual normalized Gibbs density: after a bounded unitary implementing the stated symplectic transformation, its number-basis weights are ∏j(1−e−mj)e−mjnj\prod_j(1-e^{-m_j})e^{-m_j n_j}∏j​(1−e−mj​)e−mj​nj​. This operator meaning is part of the conclusion, alongside the covariance normal form.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P46
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Intakenext100.P46

open scoped BigOperators ENNReal ComplexOrder Matrix

attribute [local instance] Classical.propDecidable Classical.decEq

open QAlgorithms.RealRegisters (Hilbert)

open QAlgorithms.Papers.Independent.Dependency50.P33 (State IsSeparable)

open QAlgorithms.Papers.Independent.Dependency50.P13
  (finiteRelativeEntropy compression entropyWeight)

open QAlgorithms.Papers.Independent.Claimed50.P47
  (Occupation bipartiteCharacteristic bipartiteWeyl gaussianStateCharacteristic)

-- These accepted definitions have the same completed Hilbert-space meaning.

open QAlgorithms.Papers.Independent.Next46.P13 (coordinateBasis)

open QAlgorithms.Papers.Independent.Frontier18.P18 (operatorCoeff boundedOperator)

open QAlgorithms.Papers.Independent.Claimed50.P41 (IsDiagonal)

/-- Chen, PDF pp.2–3, (7)–(11),(14): construct S from a real q-orthogonal
eigenbasis of p*q, including repeated eigenvalues. Printed: choose c_jj>0.
Changed: leave signs free and choose a q-orthogonal basis within each repeated
eigenspace. Why: a designated coordinate may vanish, and sign does not affect
the constructed covariance or EM. Diagonalization includes pure modes; the conditional finite-EM clause does not. -/
theorem qp_conversion {I : Type} [Fintype I] [Nonempty I]
    (q p : Matrix I I ℝ) (hphysical : IsPhysicalCovariance (Matrix.fromBlocks q 0 0 p)) :
    ∃ (g : I → ℝ) (C Sq : Matrix I I ℝ),
      (∀ j, 1 / 2 ≤ g j) ∧
      (∀ j k, ∑ l, (p * q) k l * C l j = (g j) ^ 2 * C k j) ∧
      (∀ i j, (∑ k, ∑ l, C k i * q k l * C l j) =
        if i = j then g j else 0) ∧
      Sq = q * C * Matrix.diagonal (fun j => (g j)⁻¹) ∧
      Sq.transpose⁻¹ = C ∧
      let S := Matrix.fromBlocks Sq 0 0 Sq.transpose⁻¹
      IsSymplecticMatrix S ∧
        Matrix.fromBlocks q 0 0 p = S * normalDiagonal g * S.transpose ∧
        ((∀ j, 1 / 2 < g j) →
          GaussianEM (Matrix.fromBlocks q 0 0 p)
            (S.transpose⁻¹ * normalDiagonal (modeEM ∘ g) * S⁻¹)) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P46

end
Source
Gaussian relative entropy of entanglement, arXiv:quant-ph/0402109v3, PDF p. 3, §2, Eqs. (7)–(11),(14): q–p-decorrelated CM to EM
Read-back

What the Lean code literally says, in plain math · gpt-6.1-sol

For every finite nonempty type III and real I×II\times II×I matrices q,pq,pq,p, suppose that Γ=(q00p)\Gamma=\begin{pmatrix}q&0\\0&p\end{pmatrix}Γ=(q0​0p​) is symmetric and Γ−i2J\Gamma-\tfrac i2JΓ−2i​J is positive semidefinite, where J=(0I−I0)J=\begin{pmatrix}0&I\\-I&0\end{pmatrix}J=(0−I​I0​) and real matrices are complexified in this inequality. Then there exist g:I→Rg:I\to\mathbb Rg:I→R and real matrices C,SqC,S_qC,Sq​ such that

gj≥12for every j,∑l(pq)klClj=gj2Ckjfor every j,k,∑k∑lCkiqklClj={gji=j,0i≠j,Sq=qC diag⁡(gj−1),Sq−T=C.\begin{gathered} g_j\ge\tfrac12\quad\text{for every }j,\qquad \sum_l(pq)_{kl}C_{lj}=g_j^2C_{kj}\quad\text{for every }j,k,\\ \sum_k\sum_lC_{ki}q_{kl}C_{lj} =\begin{cases}g_j&i=j,\\0&i\ne j,\end{cases} \qquad S_q=qC\,\operatorname{diag}(g_j^{-1}),\qquad S_q^{-T}=C. \end{gathered}gj​≥21​for every j,l∑​(pq)kl​Clj​=gj2​Ckj​for every j,k,k∑​l∑​Cki​qkl​Clj​={gj​0​i=j,i=j,​Sq​=qCdiag(gj−1​),Sq−T​=C.​

Writing D(g)=diag⁡(diag⁡g,diag⁡g)D(g)=\operatorname{diag}(\operatorname{diag}g,\operatorname{diag}g)D(g)=diag(diagg,diagg) and S=diag⁡(Sq,Sq−T)S=\operatorname{diag}(S_q,S_q^{-T})S=diag(Sq​,Sq−T​), the matrix SSS is invertible, satisfies SJST=JSJS^T=JSJST=J, and obeys Γ=SD(g)ST\Gamma=SD(g)S^TΓ=SD(g)ST. Moreover, if every gj>12g_j>\tfrac12gj​>21​, the pair

(Γ, M=S−TD(e∘g)S−1),e(x)=log⁡2x+12x−1,\left(\Gamma,\ M=S^{-T}D(e\circ g)S^{-1}\right), \qquad e(x)=\log\frac{2x+1}{2x-1},(Γ, M=S−TD(e∘g)S−1),e(x)=log2x−12x+1​,

has the following property: it admits a representation Γ=TD(h)TT\Gamma=TD(h)T^TΓ=TD(h)TT, M=T−TD(e∘h)T−1M=T^{-T}D(e\circ h)T^{-1}M=T−TD(e∘h)T−1 for some invertible TTT satisfying TJTT=JTJT^T=JTJTT=J and some hj>12h_j>\tfrac12hj​>21​, and every centered Gaussian density matrix RRR with covariance Γ\GammaΓ has a finite exponential representation with matrix MMM. Here a density matrix on I→NI\to\mathbb NI→N has coefficients Ruv=∑rar(u)br(v)‾R_{uv}=\sum_r a_r(u)\overline{b_r(v)}Ruv​=∑r​ar​(u)br​(v)​, with square-summable vectors ar,bra_r,b_rar​,br​ and ∑r∥ar∥∥br∥<∞\sum_r\|a_r\|\|b_r\|<\infty∑r​∥ar​∥∥br​∥<∞, real nonnegative finite quadratic forms, and a summable real diagonal with sum one. Its centered Gaussian condition requires that Γ\GammaΓ satisfy the stated physical condition and that every inner and outer sum in its characteristic be summable, with

∑u∑vRuv(Wξ)vu=e−ξTJTΓJξ/2for every ξ,\sum_u\sum_vR_{uv}(W_\xi)_{vu} =e^{-\xi^TJ^T\Gamma J\xi/2}\quad\text{for every }\xi,u∑​v∑​Ruv​(Wξ​)vu​=e−ξTJTΓJξ/2for every ξ,

where

(Wξ)uv=∏j∈ID ⁣(−ξq,j+iξp,j2)ujvj,D(z)rs=e−∣z∣2/2r!s!∑l=0min⁡(r,s)zr−l(−z‾)s−ll!(r−l)!(s−l)!.(W_\xi)_{uv} =\prod_{j\in I}D\!\left(-\frac{\xi_{q,j}+i\xi_{p,j}}{\sqrt2}\right)_{u_jv_j}, \qquad D(z)_{rs}=e^{-|z|^2/2}\sqrt{r!s!} \sum_{l=0}^{\min(r,s)} \frac{z^{r-l}(-\overline z)^{s-l}}{l!(r-l)!(s-l)!}.(Wξ​)uv​=j∈I∏​D(−2​ξq,j​+iξp,j​​)uj​vj​​,D(z)rs​=e−∣z∣2/2r!s!​l=0∑min(r,s)​l!(r−l)!(s−l)!zr−l(−z)s−l​.

The finite exponential representation means that there exist an invertible T0T_0T0​ satisfying T0JT0T=JT_0JT_0^T=JT0​JT0T​=J, positive numbers mjm_jmj​, and a unitary UUU on ℓ2(I→N)\ell^2(I\to\mathbb N)ℓ2(I→N) such that

CU−1(Wξ)=W−JT0TJξ,R=CU(Km),M=T0−TD(m)T0−1,\mathcal C_{U^{-1}}(W_\xi)=W_{-JT_0^TJ\xi},\qquad R=\mathcal C_U(K_m),\qquad M=T_0^{-T}D(m)T_0^{-1},CU−1​(Wξ​)=W−JT0T​Jξ​,R=CU​(Km​),M=T0−T​D(m)T0−1​,

where (Km)uv=0(K_m)_{uv}=0(Km​)uv​=0 off the diagonal, (Km)uu=∏j(1−e−mj)e−mjuj(K_m)_{uu}=\prod_j(1-e^{-m_j})e^{-m_ju_j}(Km​)uu​=∏j​(1−e−mj​)e−mj​uj​, and CU(K)uv=∑aUua∑bKabUvb‾\mathcal C_U(K)_{uv}=\sum_aU_{ua}\sum_bK_{ab}\overline{U_{vb}}CU​(K)uv​=∑a​Uua​∑b​Kab​Uvb​​, using Uua=(Uδa)uU_{ua}=(U\delta_a)_uUua​=(Uδa​)u​. Values gj=12g_j=\tfrac12gj​=21​ are allowed in the unconditional conclusions; the additional exponential-representation conclusion has the stated strict-inequality hypothesis.

Human review
  • Endorsed by Shuze Chen · Oct 31, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 31, 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