Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Unnumbered Eqs.(87)–(88): conditional exact singlet bound

Open
QAlgorithms.Papers.Independent.Intakenext100.P42.exact_distillation_bound

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

prove2quantum

All local registers are nonzero and finite-dimensional. Density matrices are positive semidefinite with trace one. Entropies use natural logarithms, 0ln⁡0=00\ln0=00ln0=0, and Umegaki relative entropy S(σ∥ρ)S(\sigma\Vert\rho)S(σ∥ρ) is +∞+\infty+∞ on failure of support inclusion. The relative entropy of entanglement is ER(σ)=inf⁡ρ∈SepS(σ∥ρ)E_R(\sigma)=\inf_{\rho\in\mathrm{Sep}}S(\sigma\Vert\rho)ER​(σ)=infρ∈Sep​S(σ∥ρ), where separability is across the specified Alice:Bob cut.

Let n≥1n\ge1n≥1, 0≤m≤n0\le m\le n0≤m≤n, and let σ\sigmaσ be a density matrix. Assume ER(σ⊗n)=nER(σ)E_R(\sigma^{\otimes n})=nE_R(\sigma)ER​(σ⊗n)=nER​(σ) across the grouped cut. Suppose a complete product-Kraus channel sends the entire input σ⊗n\sigma^{\otimes n}σ⊗n exactly to ∣Φm⟩⟨Φm∣⊗ω|\Phi_m\rangle\langle\Phi_m|\otimes\omega∣Φm​⟩⟨Φm​∣⊗ω, where ω\omegaω is a separable normalized residual state. Then

mln⁡2≤nER(σ).m\ln2\le nE_R(\sigma).mln2≤nER​(σ).

The channel may depend on σ\sigmaσ. This is the conditional exact fixed-output idealization, with no dropped success probability or approximate asymptotic capacity claim.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P42
Formal statement
noncomputable section

open scoped Matrix ComplexConjugate ComplexOrder ENNReal Topology

namespace QAlgorithms.Papers.Independent.Intakenext100.P42

open QAlgorithms QAlgorithms.Entanglement

open QAlgorithms.Papers.Independent.Dependency50.P21 (relativeEntropy)

open QAlgorithms.Papers.Independent.Information100.P56 (entropy)

open QAlgorithms.Papers.Independent.Information100.P04 (copies)

open QAlgorithms.Papers.Independent.Bridge50.P49 (groupedProduct)

attribute [local instance] Classical.propDecidable Classical.decEq

/-- PDF pp.30–32, Eqs.(87)–(88): conditional exact fixed-output singlet bound.
Additivity for the input copies is assumed, and the COMPLETE map produces the exact
Bell target with separable residual output. In nats the bound is m ln 2, not m. -/
theorem exact_distillation_bound {I J A B K : Type}
    [Fintype I] [Fintype J] [Fintype A] [Fintype B] [Fintype K]
    [DecidableEq I] [DecidableEq J] [DecidableEq A] [DecidableEq B]
    [Nonempty I] [Nonempty J] [Nonempty A] [Nonempty B]
    (sigma : Matrix (I × J) (I × J) ℂ) (hsigma : IsDensityMatrix sigma)
    (n m : ℕ) (hn : 1 ≤ n) (hm : m ≤ n)
    (hadd : relativeEntanglementNats (copies sigma n) = (n : EReal) * relativeEntanglementNats sigma)
    (residual : Matrix (A × B) (A × B) ℂ) (hresidual : IsSeparableDensity residual)
    (L : K → Matrix (Qubits m × A) (Fin n → I) ℂ)
    (R : K → Matrix (Qubits m × B) (Fin n → J) ℂ)
    (hcomplete : ProductComplete L R)
    (hout : krausApply (fun k => Matrix.kronecker (L k) (R k)) (copies sigma n) =
      groupedProduct (pureDensity (maxEntangled m)) residual) :
    ((m : ℝ) * Real.log 2 : EReal) ≤ (n : EReal) * relativeEntanglementNats sigma := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P42

end
Source
Entanglement Measures and Purification Procedures, arXiv:quant-ph/9707035v2, PDF p. 30, Unnumbered Eqs.(87)–(88): conditional exact singlet bound
Read-back

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

Let I,J,A,BI,J,A,BI,J,A,B be finite nonempty types with decidable equality, and let KKK be any finite type, with no nonemptiness or decidable-equality assumption on KKK. Let σ\sigmaσ be a positive semidefinite trace-one matrix on I×JI\times JI×J, let n,mn,mn,m be natural numbers with 1≤n1\le n1≤n and m≤nm\le nm≤n, and define its grouped nnn-copy matrix by (σ[n])(x,y),(x′,y′)=∏i=0n−1σ(xi,yi),(xi′,yi′)(\sigma^{[n]})_{(x,y),(x',y')}=\prod_{i=0}^{n-1}\sigma_{(x_i,y_i),(x'_i,y'_i)}(σ[n])(x,y),(x′,y′)​=∏i=0n−1​σ(xi​,yi​),(xi′​,yi′​)​. For a bipartite matrix XXX, let E(X)E(X)E(X) be the extended-real infimum of D(X∥Y)D(X\|Y)D(X∥Y) over separable density matrices YYY, meaning finite convex combinations of products of positive semidefinite trace-one matrices; here D(X∥Y)=Re⁡tr⁡[Xlog⁡X−Xlog⁡Y]D(X\|Y)=\operatorname{Re}\operatorname{tr}[X\log X-X\log Y]D(X∥Y)=Retr[XlogX−XlogY] when ker⁡Y⊆ker⁡X\ker Y\subseteq\ker XkerY⊆kerX, and +∞+\infty+∞ otherwise, with natural logarithms defined by functional calculus and log⁡0=0\log0=0log0=0. Assume explicitly that E(σ[n])=nE(σ)E(\sigma^{[n]})=nE(\sigma)E(σ[n])=nE(σ). Let η\etaη be a separable density matrix on A×BA\times BA×B, and let Lk:CIn→CQm×AL_k:\mathbb C^{I^n}\to\mathbb C^{Q_m\times A}Lk​:CIn→CQm​×A and Rk:CJn→CQm×BR_k:\mathbb C^{J^n}\to\mathbb C^{Q_m\times B}Rk​:CJn→CQm​×B be matrices indexed by k∈Kk\in Kk∈K, where Qm={0,1}mQ_m=\{0,1\}^mQm​={0,1}m, such that

∑k∈K(Lk⊗Rk)†(Lk⊗Rk)=I,∑k∈K(Lk⊗Rk)σ[n](Lk⊗Rk)†=∣Ωm⟩⟨Ωm∣⊗~η,\sum_{k\in K}(L_k\otimes R_k)^\dagger(L_k\otimes R_k)=I, \qquad \sum_{k\in K}(L_k\otimes R_k)\sigma^{[n]}(L_k\otimes R_k)^\dagger =|\Omega_m\rangle\langle\Omega_m|\mathbin{\widetilde\otimes}\eta,k∈K∑​(Lk​⊗Rk​)†(Lk​⊗Rk​)=I,k∈K∑​(Lk​⊗Rk​)σ[n](Lk​⊗Rk​)†=∣Ωm​⟩⟨Ωm​∣⊗​η,

where ∣Ωm⟩=2−m/2∑z∈Qm∣z,z⟩|\Omega_m\rangle=2^{-m/2}\sum_{z\in Q_m}|z,z\rangle∣Ωm​⟩=2−m/2∑z∈Qm​​∣z,z⟩ and the grouped product on the right has entry ∣Ωm⟩⟨Ωm∣(z,w),(z′,w′)η(a,b),(a′,b′)|\Omega_m\rangle\langle\Omega_m|_{(z,w),(z',w')}\eta_{(a,b),(a',b')}∣Ωm​⟩⟨Ωm​∣(z,w),(z′,w′)​η(a,b),(a′,b′)​ at row ((z,a),(w,b))((z,a),(w,b))((z,a),(w,b)) and column ((z′,a′),(w′,b′))((z',a'),(w',b'))((z′,a′),(w′,b′)). Then

mlog⁡2≤nE(σ)m\log2\le nE(\sigma)mlog2≤nE(σ)

in the extended reals. The case m=0m=0m=0 is included. Although KKK may formally be empty, the completeness equation cannot hold for empty KKK because the input space is nonzero.

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