Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proposition 1: multipartite entanglement minimum

Open
QAlgorithms.Papers.Independent.Support50.P14.multipartiteEntanglementMinimum

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

prove2quantum

Consider M≥1M\ge1M≥1 parties, each holding an arbitrary positive finite number of full oscillator modes. Let EEE be a compatible real-valued functional family that is continuous in trace norm, strongly superadditive with equality on product states, and invariant under independent local unitaries. In particular these are the properties required of the entanglement measure in Proposition 1; local-unitary invariance is part of that convention. If ρ\rhoρ has finite complete moments and ρG\rho_GρG​ is any Gaussian density with the same symmetric covariance, then

E(ρG)≤E(ρ).E(\rho_G)\le E(\rho).E(ρG​)≤E(ρ).

The first moments may differ. The statement displays the functional properties actually used by the proposition, so it applies to every entanglement measure satisfying its hypotheses without selecting one named measure.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P14
Formal statement
noncomputable section

open Filter

open scoped BigOperators ENNReal Topology

attribute [local instance] Classical.propDecidable Classical.decEq

namespace QAlgorithms.Papers.Independent.Support50.P14

open QAlgorithms.RealRegisters (Hilbert)

open QAlgorithms.Papers.Independent.Dependency50.P33
  (State IsDensity IsTraceClass traceSize)

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

open QAlgorithms.Papers.Independent.Frontier30.P14
  (unitaryCoeff conjugateKernel boundedUnitaryConjugation)

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

open QAlgorithms.Papers.Independent.Claimed50.P47 (Occupation)

open QAlgorithms.Papers.Independent.Gap48.P20 (Copies)

/-- Proposition 1, PDF p.3: arbitrary finite nonempty mode groups at M parties.
The entanglement-measure convention is used through its local-unitary invariance;
the other requirements are trace continuity and strong super-additivity with
product equality. The conclusion is valid for every functional with these
properties, hence for every entanglement measure satisfying the stated axioms.
Only covariance is shared; d and dG are independent. -/
theorem multipartiteEntanglementMinimum {M : ℕ} (hM : 0 < M) (E : MultipartiteOscillatorFunctional M)
    (hcont : OscillatorTraceContinuous none E) (hssa : StronglySuperadditive none E)
    (hentanglementLocalInvariance : LocallyInvariant none E)
    {I : Type} [Fintype I] (owner : I → Fin M) (howner : Function.Surjective owner)
    (rho rhoG : State (Occupation I)) (d dG : InterleavedQuadrature I → ℝ) (gamma : OscillatorCovariance I)
    (hmoments : CompleteMoments rho.coefficients d gamma)
    (hgaussian : GaussianAt rhoG.coefficients dG gamma) :
    E I owner rhoG.coefficients ≤ E I owner rho.coefficients := by sorry

end QAlgorithms.Papers.Independent.Support50.P14

end
Source
Extremality of Gaussian quantum states, arXiv:quant-ph/0509154v1, PDF p. 3, Proposition 1: multipartite entanglement minimum
Read-back

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

Let MMM be a positive natural number, and let EEE assign a real number E(I,o,L)E(I,o,L)E(I,o,L) to every finite type III, every map o:I→{0,…,M−1}o:I\to\{0,\ldots,M-1\}o:I→{0,…,M−1}, and every complex kernel LLL on NI=NI\mathcal N_I=\mathbb N^INI​=NI. For any label set Λ\LambdaΛ, call a kernel LLL a density kernel when it admits an expansion Luv=∑rar(u)br(v)‾L_{uv}=\sum_r a_r(u)\overline{b_r(v)}Luv​=∑r​ar​(u)br​(v)​ with ar,br∈ℓ2(Λ;C)a_r,b_r\in\ell^2(\Lambda;\mathbb C)ar​,br​∈ℓ2(Λ;C) and ∑r∥ar∥2∥br∥2<∞\sum_r\|a_r\|_2\|b_r\|_2<\infty∑r​∥ar​∥2​∥br​∥2​<∞, every finite principal submatrix is positive semidefinite, and its real diagonal is summable with sum 111. Define

T(L)=inf⁡(ar,br)∑r∥ar∥2∥br∥2,T(L)=\inf_{(a_r,b_r)} \sum_r\|a_r\|_2\|b_r\|_2,T(L)=(ar​,br​)inf​r∑​∥ar​∥2​∥br​∥2​,

as an extended nonnegative real, where the infimum ranges over exactly those expansions; if none exists, the infimum is +∞+\infty+∞. Assume the following properties of EEE on all density kernels, without an energy cap. First, for every finite III, ownership map ooo, density kernel LLL, and ϵ>0\epsilon>0ϵ>0, there is δ>0\delta>0δ>0 such that every density kernel L′L'L′ on the same labels with T(L′−L)<δT(L'-L)<\deltaT(L′−L)<δ satisfies ∣E(I,o,L′)−E(I,o,L)∣<ϵ|E(I,o,L')-E(I,o,L)|<\epsilon∣E(I,o,L′)−E(I,o,L)∣<ϵ. Second, for any finite A,BA,BA,B, ownership maps a,ba,ba,b, and kernel LLL on NA⊔B\mathbb N^{A\sqcup B}NA⊔B, define

LA(u,v)=∑w∈NBLu⊔w,v⊔w,LB(u,v)=∑w∈NALw⊔u,w⊔v.L_A(u,v)=\sum_{w\in\mathbb N^B}L_{u\sqcup w,v\sqcup w}, \qquad L_B(u,v)=\sum_{w\in\mathbb N^A}L_{w\sqcup u,w\sqcup v}.LA​(u,v)=w∈NB∑​Lu⊔w,v⊔w​,LB​(u,v)=w∈NA∑​Lw⊔u,w⊔v​.

Whenever L,LA,LBL,L_A,L_BL,LA​,LB​ are density kernels, assume

E(A,a,LA)+E(B,b,LB)≤E(A⊔B,a⊔b,L).E(A,a,L_A)+E(B,b,L_B)\le E(A\sqcup B,a\sqcup b,L).E(A,a,LA​)+E(B,b,LB​)≤E(A⊔B,a⊔b,L).

For kernels LLL on NA\mathbb N^ANA and KKK on NB\mathbb N^BNB, put (L⊗K)uv=Lu∣A,v∣AKu∣B,v∣B(L\otimes K)_{uv}=L_{u|_A,v|_A}K_{u|_B,v|_B}(L⊗K)uv​=Lu∣A​,v∣A​​Ku∣B​,v∣B​​; whenever L,K,L⊗KL,K,L\otimes KL,K,L⊗K are density kernels, assume

E(A⊔B,a⊔b,L⊗K)=E(A,a,L)+E(B,b,K).E(A\sqcup B,a\sqcup b,L\otimes K)=E(A,a,L)+E(B,b,K).E(A⊔B,a⊔b,L⊗K)=E(A,a,L)+E(B,b,K).

Third, for every finite III, ownership map ooo, and family of complex linear surjective isometries UpU_pUp​ on ℓ2(NIp;C)\ell^2(\mathbb N^{I_p};\mathbb C)ℓ2(NIp​;C), where Ip={j∈I:o(j)=p}I_p=\{j\in I:o(j)=p\}Ip​={j∈I:o(j)=p}, write eve_vev​ for the coordinate unit vector and define

KU(u,v)=∏p=0M−1(Upev∣Ip)(u∣Ip),CU(L)uv=∑xKU(u,x)(∑yLxyKU(v,y)‾).K_U(u,v)=\prod_{p=0}^{M-1}(U_pe_{v|_{I_p}})(u|_{I_p}), \qquad C_U(L)_{uv}=\sum_x K_U(u,x) \left(\sum_y L_{xy}\overline{K_U(v,y)}\right).KU​(u,v)=p=0∏M−1​(Up​ev∣Ip​​​)(u∣Ip​​),CU​(L)uv​=x∑​KU​(u,x)(y∑​Lxy​KU​(v,y)​).

Whenever L,CU(L)L,C_U(L)L,CU​(L) are density kernels, assume E(I,o,CU(L))=E(I,o,L)E(I,o,C_U(L))=E(I,o,L)E(I,o,CU​(L))=E(I,o,L). Also assume invariance under every relabeling bijection e:I→Je:I\to Je:I→J: if LLL and Luve=Lu∘e,v∘eL^e_{uv}=L_{u\circ e,v\circ e}Luve​=Lu∘e,v∘e​ are density kernels and o:J→{0,…,M−1}o:J\to\{0,\ldots,M-1\}o:J→{0,…,M−1}, then E(J,o,Le)=E(I,o∘e,L)E(J,o,L^e)=E(I,o\circ e,L)E(J,o,Le)=E(I,o∘e,L). These assumptions quantify over empty finite mode sets as well, and their ownership maps need not be surjective. Now let III be finite, let o:I→{0,…,M−1}o:I\to\{0,\ldots,M-1\}o:I→{0,…,M−1} be surjective, and let ρ,ρG\rho,\rho_Gρ,ρG​ be states with density kernels L,LGL,L_GL,LG​. Thus the present III is nonempty and every party owns at least one mode. Let d,dG:I×{0,1}→Rd,d_G:I\times\{0,1\}\to\mathbb Rd,dG​:I×{0,1}→R and let γ\gammaγ be a real matrix on that quadrature set. Define

Dz(m,n)=e−∣z∣2/2m! n!∑j=0min⁡(m,n)zm−j(−z‾)n−jj!(m−j)!(n−j)!,Wξ(u,v)=∏i∈ID(−ξi,1+iξi,0)/2(ui,vi),D_z(m,n)=e^{-|z|^2/2}\sqrt{m!\,n!} \sum_{j=0}^{\min(m,n)} \frac{z^{m-j}(-\overline z)^{n-j}}{j!(m-j)!(n-j)!}, \quad W_\xi(u,v)=\prod_{i\in I} D_{(-\xi_{i,1}+i\xi_{i,0})/\sqrt2}(u_i,v_i),Dz​(m,n)=e−∣z∣2/2m!n!​j=0∑min(m,n)​j!(m−j)!(n−j)!zm−j(−z)n−j​,Wξ​(u,v)=i∈I∏​D(−ξi,1​+iξi,0​)/2​​(ui​,vi​), χK(ξ)=∑u∑vKuvWξ(v,u),me(ξ)=∑aξaea,Vγ(ξ)=∑a,bξaγabξb.\chi_K(\xi)=\sum_u\sum_v K_{uv}W_\xi(v,u),\qquad m_e(\xi)=\sum_a\xi_ae_a,\qquad V_\gamma(\xi)=\sum_{a,b}\xi_a\gamma_{ab}\xi_b.χK​(ξ)=u∑​v∑​Kuv​Wξ​(v,u),me​(ξ)=a∑​ξa​ea​,Vγ​(ξ)=a,b∑​ξa​γab​ξb​.

Assume γT=γ\gamma^\mathsf T=\gammaγT=γ; that both LLL and LGL_GLG​ have finite energy ∑u[∑iui]+[Re⁡Kuu]+\sum_u[\sum_i u_i]_+[\operatorname{Re}K_{uu}]_+∑u​[∑i​ui​]+​[ReKuu​]+​, where [x]+=max⁡(x,0)[x]_+=\max(x,0)[x]+​=max(x,0); and that for each of these kernels and every ξ\xiξ, all inner characteristic series and the outer series of inner sums are summable. Assume t↦χL(tξ)t\mapsto\chi_L(t\xi)t↦χL​(tξ) has derivative i md(ξ)i\,m_d(\xi)imd​(ξ) at 000, and its derivative function has derivative −Vγ(ξ)/2−md(ξ)2-V_\gamma(\xi)/2-m_d(\xi)^2−Vγ​(ξ)/2−md​(ξ)2 there; assume the corresponding derivative statements for LG,dGL_G,d_GLG​,dG​, and additionally

χLG(ξ)=exp⁡ ⁣(−Vγ(ξ)4+i mdG(ξ))for every ξ.\chi_{L_G}(\xi)= \exp\!\left(-\frac{V_\gamma(\xi)}4+i\,m_{d_G}(\xi)\right) \quad\text{for every }\xi.χLG​​(ξ)=exp(−4Vγ​(ξ)​+imdG​​(ξ))for every ξ.

The vectors ddd and dGd_GdG​ need not agree. Under all these hypotheses, the conclusion is

E(I,o,LG)≤E(I,o,L).E(I,o,L_G)\le E(I,o,L).E(I,o,LG​)≤E(I,o,L).
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