Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Reflected Gaussian purification

Open
QAlgorithms.Papers.Independent.Support50.P29.gaussianPurification

by Goku · Oct 23, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let s≥1s\ge1s≥1, ℏ>0\hbar>0ℏ>0, and Δ=ℏ[0I−I0]\Delta=\hbar\begin{bmatrix}0&I\\-I&0\end{bmatrix}Δ=ℏ[0−I​I0​]. Let ρ\rhoρ be a normalized centered Gaussian state of a regular irreducible canonical Weyl system on a nonzero complete separable Hilbert space, with real symmetric covariance α\alphaα satisfying α−iΔ/2≥0\alpha-i\Delta/2\ge0α−iΔ/2≥0. Choose a second regular irreducible oscillator system and reflect its reference coordinates by exchanging qqq and ppp, so its CCR form is −Δ-\Delta−Δ. Set

S=−(Δ−1α)2−I/4,α12=[αΔS−ΔSα],Δ12=diag⁡(Δ,−Δ).S=\sqrt{-(\Delta^{-1}\alpha)^2-I/4},\qquad \alpha_{12}=\begin{bmatrix}\alpha&\Delta S\\-\Delta S&\alpha\end{bmatrix},\qquad \Delta_{12}=\operatorname{diag}(\Delta,-\Delta).S=−(Δ−1α)2−I/4​,α12​=[α−ΔS​ΔSα​],Δ12​=diag(Δ,−Δ).

The square root is the nonnegative branch of similarity spectral calculus. There exists a normalized Gaussian density ρ12\rho_{12}ρ12​ on the completed tensor product, with characteristic exp⁡(−zTα12z/2)\exp(-z^T\alpha_{12}z/2)exp(−zTα12​z/2), such that

Tr⁡2ρ12=ρ,ρ12=∣ψ⟩⟨ψ∣(∥ψ∥=1),(Δ12−1α12)2=−I/4.\operatorname{Tr}_2\rho_{12}=\rho,\qquad \rho_{12}=|\psi\rangle\langle\psi|\quad(\|\psi\|=1),\qquad (\Delta_{12}^{-1}\alpha_{12})^2=-I/4.Tr2​ρ12​=ρ,ρ12​=∣ψ⟩⟨ψ∣(∥ψ∥=1),(Δ12−1​α12​)2=−I/4.

Thus the displayed covariance specifies an actual pure state and the correct marginal, not merely a physical covariance witness. The paper acknowledges the construction's provenance in Holevo [4].

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P29
Formal statement
noncomputable section

open Filter

open scoped BigOperators ENNReal Topology ComplexOrder

attribute [local instance] Classical.propDecidable Classical.decEq

namespace QAlgorithms.Papers.Independent.Support50.P29

open QAlgorithms.RealRegisters (Hilbert)

open QAlgorithms.Papers.Independent.Dependency50.P10 (Phase sigma complexify)

open QAlgorithms.Papers.Independent.Dependency50.P33
  (State IsDensity rankOne productCoeff tensorApply IsTraceClassChannel)

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

open QAlgorithms.Papers.Independent.Frontier18.P07 (extendedEntropyNats)

open QAlgorithms.Papers.Independent.Frontier30.P10 (WeylSystem)

open QAlgorithms.Papers.Independent.Frontier30.P14 (partialRight PureDensity)

open QAlgorithms.Papers.Independent.Claimed50.P35
  (tracePair thermalInput thermalPurificationDensity vacuumChannel exchangeMatrix RealizedDensity)

open QAlgorithms.Papers.Independent.Dependency50.P39 (displacementMatrix)

open QAlgorithms.Papers.Independent.Information100.P49 (thermalEntropy)

/-- Holevo quant-ph/9809022v1, PDF pp.6-7, §3: reflected-reference
Gaussian purification with the displayed covariance, actual partial trace,
and norm-one rank-one purity. The source acknowledges construction [4]. -/
theorem gaussianPurification {I J : Type} [Countable I] [Countable J]
    [Nonempty I] [Nonempty J] (s : ℕ) (hs : 0 < s)
    (hbar : ℝ) (hhbar : 0 < hbar)
    (W : WeylSystem (Hilbert I) s) (V : WeylSystem (Hilbert J) s)
    (rho : State I) (alpha : Matrix (Phase s) (Phase s) ℝ)
    (hrho : CenteredGaussianWithCovariance hbar W rho alpha) :
    (∃ rho12 : State (I × J),
      PureDensity rho12.coefficients ∧
      partialRight rho12.coefficients = rho.coefficients ∧
      (∀ z, jointCharacteristic hbar W V rho12.coefficients z =
        centeredCharacteristic (purificationCovariance hbar alpha) z)) ∧
    (((reflectedDelta s hbar)⁻¹ * purificationCovariance hbar alpha) ^ 2 =
      (-1 / 4 : ℝ) • 1) := by sorry

end QAlgorithms.Papers.Independent.Support50.P29

end
Source
Sending Quantum Information with Gaussian States, arXiv:quant-ph/9809022v1, PDF p. 6, §3, unnumbered: constructed Gaussian purification with reflected CCR
Read-back

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

Let I,JI,JI,J be countable, nonempty types, let sss be a positive natural number, and let ℏ>0\hbar>0ℏ>0. For any label set LLL, write HL=ℓ2(L;C)H_L=\ell^2(L;\mathbb C)HL​=ℓ2(L;C); a state on LLL means a coefficient matrix RRR admitting an expansion Rab=∑r=0∞ur(a)vr(b)‾R_{ab}=\sum_{r=0}^\infty u_r(a)\overline{v_r(b)}Rab​=∑r=0∞​ur​(a)vr​(b)​ with ur,vr∈HLu_r,v_r\in H_Lur​,vr​∈HL​ and ∑r∥ur∥∥vr∥<∞\sum_r\|u_r\|\|v_r\|<\infty∑r​∥ur​∥∥vr​∥<∞, such that every finite quadratic form ∑a,b∈Sxa‾Rabxb\sum_{a,b\in S}\overline{x_a}R_{ab}x_b∑a,b∈S​xa​​Rab​xb​ is real and nonnegative and the real diagonal entries are summable with sum 111. Let FFF consist of sss position indices and sss momentum indices. Suppose WWW on HIH_IHI​ and VVV on HJH_JHJ​ are families of complex linear isometric equivalences indexed by RF\mathbb R^FRF, each equal to the identity at zero, continuous on each vector, irreducible in the sense that its only closed invariant complex linear subspaces are {0}\{0\}{0} and the whole space, and satisfying

W(z)W(w)=exp⁡ ⁣(i2∑j=1s(zpjwqj−zqjwpj))W(z+w),W(z)W(w)= \exp\!\left(\frac{i}{2}\sum_{j=1}^s (z_{p_j}w_{q_j}-z_{q_j}w_{p_j})\right)W(z+w),W(z)W(w)=exp(2i​j=1∑s​(zpj​​wqj​​−zqj​​wpj​​))W(z+w),

with the same identity for VVV. Put Δ=ℏ(01s−1s0)\Delta=\hbar\begin{pmatrix}0&1_s\\-1_s&0\end{pmatrix}Δ=ℏ(0−1s​​1s​0​), and let ρ\rhoρ be a state on III, with coefficient matrix RRR. Suppose α\alphaα is a real symmetric 2s×2s2s\times2s2s×2s matrix for which α−i2Δ\alpha-\frac{i}{2}\Deltaα−2i​Δ is positive semidefinite and

∑a,b∈IRab(W(ℏ z)ea)b=e−zTαz/2for every z∈RF.\sum_{a,b\in I}R_{ab}\bigl(W(\sqrt{\hbar}\,z)e_a\bigr)_b =e^{-z^{\mathsf T}\alpha z/2} \quad\text{for every }z\in\mathbb R^F.a,b∈I∑​Rab​(W(ℏ​z)ea​)b​=e−zTαz/2for every z∈RF.

For a finite real matrix MMM, define S(M)\mathcal S(M)S(M) as follows: if its complexification admits an invertible diagonalization M=Qdiag⁡(d)Q−1M=Q\operatorname{diag}(d)Q^{-1}M=Qdiag(d)Q−1, choose one and take the entrywise real part of Qdiag⁡(Re⁡dj)Q−1Q\operatorname{diag}(\sqrt{\operatorname{Re}d_j})Q^{-1}Qdiag(Redj​​)Q−1; otherwise set S(M)=0\mathcal S(M)=0S(M)=0. Here the real square root is total and equals zero on negative inputs. Set

A=Δ−1α,B=Δ S ⁣(−A2−141),Γ=(αB−Bα),Δref=(Δ00−Δ).A=\Delta^{-1}\alpha,\qquad B=\Delta\,\mathcal S\!\left(-A^2-\tfrac14 1\right),\qquad \Gamma=\begin{pmatrix}\alpha&B\\-B&\alpha\end{pmatrix}, \qquad \Delta_{\mathrm{ref}}=\begin{pmatrix}\Delta&0\\0&-\Delta\end{pmatrix}.A=Δ−1α,B=ΔS(−A2−41​1),Γ=(α−B​Bα​),Δref​=(Δ0​0−Δ​).

For operators CCC on HIH_IHI​ and DDD on HJH_JHJ​, write Cab=(Ceb)aC_{ab}=(Ce_b)_aCab​=(Ceb​)a​ and Dcd=(Ded)cD_{cd}=(De_d)_cDcd​=(Ded​)c​, and define the joint characteristic of a coefficient matrix XXX at (z,w)∈RF⊕RF(z,w)\in\mathbb R^F\oplus\mathbb R^F(z,w)∈RF⊕RF to be

χX(z,w)=∑(a,c)∈I×J∑(b,d)∈I×JX(a,c),(b,d)(W(ℏ z))ba(V(ℏ Sw))dc,\chi_X(z,w)= \sum_{(a,c)\in I\times J}\sum_{(b,d)\in I\times J} X_{(a,c),(b,d)} \bigl(W(\sqrt{\hbar}\,z)\bigr)_{ba} \bigl(V(\sqrt{\hbar}\,Sw)\bigr)_{dc},χX​(z,w)=(a,c)∈I×J∑​(b,d)∈I×J∑​X(a,c),(b,d)​(W(ℏ​z))ba​(V(ℏ​Sw))dc​,

where SSS exchanges the position and momentum coordinates, (Sw)qj=wpj(Sw)_{q_j}=w_{p_j}(Sw)qj​​=wpj​​ and (Sw)pj=wqj(Sw)_{p_j}=w_{q_j}(Sw)pj​​=wqj​​. Then there exists a state ρ12\rho_{12}ρ12​ on I×JI\times JI×J, with coefficient matrix R12R_{12}R12​, which also satisfies the density conditions above and has the form R12(p,q)=u(p)u(q)‾R_{12}(p,q)=u(p)\overline{u(q)}R12​(p,q)=u(p)u(q)​ for some u∈HI×Ju\in H_{I\times J}u∈HI×J​ of norm 111, such that

∑c∈J(R12)(a,c),(b,c)=Rabfor all a,b∈I,χR12(z,w)=exp⁡ ⁣(−12(z,w)TΓ(z,w))for all z,w.\sum_{c\in J}(R_{12})_{(a,c),(b,c)}=R_{ab} \quad\text{for all }a,b\in I, \qquad \chi_{R_{12}}(z,w)= \exp\!\left(-\tfrac12(z,w)^{\mathsf T}\Gamma(z,w)\right) \quad\text{for all }z,w.c∈J∑​(R12​)(a,c),(b,c)​=Rab​for all a,b∈I,χR12​​(z,w)=exp(−21​(z,w)TΓ(z,w))for all z,w.

In addition,

(Δref−1Γ)2=−141.(\Delta_{\mathrm{ref}}^{-1}\Gamma)^2=-\tfrac14 1.(Δref−1​Γ)2=−41​1.
Human review
  • Endorsed by Shuze Chen · Oct 24, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Goku · Oct 24, 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