Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Exact extraction from a rectangular resource

Open
QAlgorithms.Papers.Independent.Intakenext100.P18.rectangular_extraction

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

prove2quantum

Let 1≤N<M1\le N<M1≤N<M, let ∣ΦN⟩=N−1/2∑i=0N−1∣i⟩A∣i⟩B|\Phi_N\rangle=N^{-1/2}\sum_{i=0}^{N-1}|i\rangle_A|i\rangle_B∣ΦN​⟩=N−1/2∑i=0N−1​∣i⟩A​∣i⟩B​ in CN⊗CM\mathbb C^N\otimes\mathbb C^MCN⊗CM, and let PPP be Bob's orthogonal projection onto his first NNN coordinates. Let τ\tauτ be a density operator whose Bob marginal obeys PTr⁡Aτ=0P\operatorname{Tr}_A\tau=0PTrA​τ=0. For 0<p≤10<p\le10<p≤1, put ρ=p∣ΦN⟩⟨ΦN∣+(1−p)τ\rho=p|\Phi_N\rangle\langle\Phi_N|+(1-p)\tauρ=p∣ΦN​⟩⟨ΦN​∣+(1−p)τ. Then ρ\rhoρ is a density, III and PPP are physical local contractions, and

(I⊗P)ρ(I⊗P)=p∣ΦN⟩⟨ΦN∣.(I\otimes P)\rho(I\otimes P)=p|\Phi_N\rangle\langle\Phi_N|.(I⊗P)ρ(I⊗P)=p∣ΦN​⟩⟨ΦN​∣.

The success probability is exactly ppp, and the conditional output is exactly the embedded plus state. In particular ρ\rhoρ is exactly noncollectively distillable to the NNN-dimensional plus state. If p<1p<1p<1, ρ\rhoρ is mixed (rank greater than one).

Only Bob's noise marginal is restricted; there is no extra condition on Alice's marginal. Strict positivity of ppp excludes the inadmissible zero-success endpoint.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P18
Formal statement
noncomputable section

open QAlgorithms MeasureTheory

open scoped Matrix ComplexConjugate ComplexOrder Matrix.Norms.L2Operator

namespace QAlgorithms.Papers.Independent.Intakenext100.P18

open QAlgorithms.Papers.Independent.Claimed50.P28 (maximallyEntangledDensity)

open QAlgorithms.Papers.Independent.Support50.P11 (choiState teleportWeyl bellKet)

open QAlgorithms.Papers.Independent.Renew46.P40 (initialJoint)

/-- PDF p.10, rectangular example: Bob projects onto the first N coordinates.
Only Bob's noise marginal must be orthogonal. Success is exactly p>0.
For strict mixtures this is a mixed-state example, not a pure-state restatement. -/
theorem rectangular_extraction (N M : ℕ) (hN : 0 < N) (hNM : N < M)
    (p : ℝ) (hp : 0 < p) (hp1 : p ≤ 1)
    (tau : Matrix (Fin N × Fin M) (Fin N × Fin M) ℂ) (htau : IsDensityMatrix tau)
    (horth : coordinateProjection N M * traceLeft tau = 0) :
    IsDensityMatrix (rectangularResource N M p tau) ∧
    IsLocalFilter (1 : Matrix (Fin N) (Fin N) ℂ) ∧
    IsLocalFilter (coordinateProjection N M) ∧
    (let omega := conjSuperop
      (Matrix.kronecker (1 : Matrix (Fin N) (Fin N) ℂ) (coordinateProjection N M))
      (rectangularResource N M p tau);
     omega = (p : ℂ) • rectangularPlus N M ∧ omega.trace.re = p ∧
     ((omega.trace.re)⁻¹ : ℂ) • omega = rectangularPlus N M) ∧
    ExactlyDistillable (rectangularResource N M p tau) N hN ∧
    (p < 1 → 1 < (rectangularResource N M p tau).rank) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P18

end
Source
General teleportation channel, singlet fraction and quasi-distillation, arXiv:quant-ph/9807091v2, PDF p. 10, §VI, unnumbered example: exact extraction from a rectangular resource
Read-back

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

For all integers 0<N<M0<N<M0<N<M, every real ppp with 0<p≤10<p\le10<p≤1, and every positive semidefinite trace-one matrix τ\tauτ on CN⊗CM\mathbb C^N\otimes\mathbb C^MCN⊗CM, let QQQ be the diagonal M×MM\times MM×M matrix with Qii=1Q_{ii}=1Qii​=1 for i<Ni<Ni<N and 000 otherwise, and suppose QTr⁡1τ=0Q\operatorname{Tr}_1\tau=0QTr1​τ=0, where (Tr⁡1τ)ab=∑iτ(i,a),(i,b)(\operatorname{Tr}_1\tau)_{ab}=\sum_i\tau_{(i,a),(i,b)}(Tr1​τ)ab​=∑i​τ(i,a),(i,b)​. Put

∣v⟩=1N∑j=0N−1∣j,j⟩,R=∣v⟩⟨v∣,ρ=pR+(1−p)τ.|v\rangle=\frac1{\sqrt N}\sum_{j=0}^{N-1}|j,j\rangle,\qquad R=|v\rangle\langle v|,\qquad \rho=pR+(1-p)\tau.∣v⟩=N​1​j=0∑N−1​∣j,j⟩,R=∣v⟩⟨v∣,ρ=pR+(1−p)τ.

Then ρ\rhoρ is positive semidefinite with trace 111, both INI_NIN​ and QQQ are local filters in the sense that their respective matrices I−A†AI-A^\dagger AI−A†A are positive semidefinite, and

ω=(IN⊗Q)ρ(IN⊗Q)†=pR,Re⁡tr⁡ω=p,ωRe⁡tr⁡ω=R.\omega=(I_N\otimes Q)\rho(I_N\otimes Q)^\dagger=pR,\qquad \operatorname{Re}\operatorname{tr}\omega=p,\qquad \frac{\omega}{\operatorname{Re}\operatorname{tr}\omega}=R.ω=(IN​⊗Q)ρ(IN​⊗Q)†=pR,Retrω=p,Retrωω​=R.

Furthermore, there exist A:CN→CNA:\mathbb C^N\to\mathbb C^NA:CN→CN and B:CM→CNB:\mathbb C^M\to\mathbb C^NB:CM→CN with IN−A†AI_N-A^\dagger AIN​−A†A and IM−B†BI_M-B^\dagger BIM​−B†B positive semidefinite such that ω~=(A⊗B)ρ(A⊗B)†\widetilde\omega=(A\otimes B)\rho(A\otimes B)^\daggerω=(A⊗B)ρ(A⊗B)† has positive real trace and ω~/Re⁡tr⁡ω~=∣ΩN⟩⟨ΩN∣\widetilde\omega/\operatorname{Re}\operatorname{tr}\widetilde\omega=|\Omega_N\rangle\langle\Omega_N|ω/Retrω=∣ΩN​⟩⟨ΩN​∣, where ∣ΩN⟩=N−1/2∑j=0N−1∣j,j⟩|\Omega_N\rangle=N^{-1/2}\sum_{j=0}^{N-1}|j,j\rangle∣ΩN​⟩=N−1/2∑j=0N−1​∣j,j⟩ now lies in CN⊗CN\mathbb C^N\otimes\mathbb C^NCN⊗CN. If p<1p<1p<1, then rank⁡ρ>1\operatorname{rank}\rho>1rankρ>1. The assumptions permit N=1N=1N=1 and p=1p=1p=1; the last rank conclusion is conditional on p<1p<1p<1.

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