Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem IV.7: For fully quantum multiple source, any quantum-Fe achievable R obeys every sum_(i∈I)Ri≥H(A(I)|A(Ic))=H(averageState)−H(averageStat

Open
QAlgorithms.Papers.Independent.Renew46.P43.entanglementSourceConverse

by Goku · Oct 22, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Consider a finite pure-letter iid multiple source, allowing correlated classical laws and entangled source letters where the model permits them. Quantum rate regions use separate local CPTP encoders and a joint CPTP decoder, with rates converging and the named average half-trace, average-fidelity or entanglement-fidelity error tending to zero. Entanglement fidelity is used only when every sender system is a full matrix algebra. For classical/hybrid senders X1,…,Xs,YX_1,\ldots,X_s,YX1​,…,Xs​,Y, conditional entropies are evaluated in the computed joint source average, the Y rate is exactly H(Y)H(Y)H(Y), and every sender subset, including empty/full, is tested.

∀J⊆[s]:∑i∈JRi≥H(ρˉ)−H(ρˉJc).\forall J\subseteq[s]:\quad\sum_{i\in J}R_i\ge H(\bar\rho)-H(\bar\rho_{J^c}).∀J⊆[s]:i∈J∑​Ri​≥H(ρˉ​)−H(ρˉ​Jc​).

This is the source result with its stated operational or information-theoretic domain.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Renew46_P43
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Renew46.P43

open scoped BigOperators Matrix ComplexConjugate ComplexOrder ENNReal Matrix.Norms.L2Operator

open Filter MeasureTheory

open QAlgorithms.Papers.Independent.Information100.P08 (IsPure)

open QAlgorithms.Papers.Independent.Information100.P06 (Below)

open QAlgorithms.Papers.Independent.Information100.P60 (copies relativeBits)

open QAlgorithms.Papers.Independent.Dependency50.P19 (count both)

open QAlgorithms.Papers.Independent.Information100.P22 (coherentInformation)

attribute [local instance] Classical.propDecidable Classical.decEq

open QAlgorithms.Papers.Independent.Claimed50.P40 (InBlocks)

attribute [local instance] QAlgorithms.Papers.Independent.Renew46.P43.matrixMeasurableSpace

/-- Theorem IV.7, PDF pp.50-51: every subset, empty and full included; no sender
product-state promise or independent sender-law assumption. -/
theorem entanglementSourceConverse {s : ℕ} {L : Type*} [Fintype L]
    (S : MultipleData s L) (hS : S.Valid)
    (hfull : ∀ i a b, (S.system i).block a = (S.system i).block b)
    (R : Fin s → ℝ) (hR : R ∈ entanglementSourceRegion S hfull) :
    ∀ J : Finset (Fin s), ∑ i ∈ J, R i ≥
      vnEntropy S.average - vnEntropy (multipleMarginal S.system S.average Jᶜ) := by sorry

end QAlgorithms.Papers.Independent.Renew46.P43

end
Source
Coding Theorems of Quantum Information Theory, arXiv:quant-ph/9907077v1, PDF p. 50, Theorem IV.7
Read-back

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

Let sss be any natural number, including 000, let LLL be a finite type, and let SSS specify, for each i∈{0,…,s−1}i\in\{0,\ldots,s-1\}i∈{0,…,s−1}, a positive system dimension did_idi​ and a block-label function bib_ibi​, together with a probability mass function ppp on LLL and joint matrices ρx\rho_xρx​. Assume that every ρx\rho_xρx​ is the outer product of a unit vector with itself, that x↦ρxx\mapsto\rho_xx↦ρx​ is injective, and that its entries vanish whenever some coordinate’s row and column block labels differ. Assume additionally that bi(a)=bi(b)b_i(a)=b_i(b)bi​(a)=bi​(b) for every party iii and every pair of its basis indices a,ba,ba,b. Put ρˉ=∑xp(x)ρx\bar\rho=\sum_xp(x)\rho_xρˉ​=∑x​p(x)ρx​, and let R:{0,…,s−1}→RR:\{0,\ldots,s-1\}\to\mathbb RR:{0,…,s−1}→R. Suppose that at every positive integer length NNN there are positive compressed dimensions Di(N)D_i(N)Di​(N), separate encoding channels Ei,NE_{i,N}Ei,N​, and a joint decoding channel DND_NDN​, with composite map TN=DN∘⨂iEi,N\mathcal T_N=D_N\circ\bigotimes_iE_{i,N}TN​=DN​∘⨂i​Ei,N​, such that log⁡2Di(N)/N→Ri\log_2D_i(N)/N\to R_ilog2​Di​(N)/N→Ri​. A channel here has a finite Kraus representation X↦∑jKjXKj†X\mapsto\sum_jK_jXK_j^\daggerX↦∑j​Kj​XKj†​ with ∑jKj†Kj=I\sum_jK_j^\dagger K_j=I∑j​Kj†​Kj​=I. The code also requires decoded positive semidefinite inputs to have zero entries whenever any output block labels differ; under the assumed equality of all labels within each system, this condition imposes no entry restrictions. For ρN=ρˉ⊗N\rho_N=\bar\rho^{\otimes N}ρN​=ρˉ​⊗N, regrouped by party, define vN(a,b)=(ρN)b,av_N(a,b)=(\sqrt{\rho_N})_{b,a}vN​(a,b)=(ρN​​)b,a​ and PN=∣vN⟩⟨vN∣P_N=|v_N\rangle\langle v_N|PN​=∣vN​⟩⟨vN​∣, and suppose

lim⁡N→∞[1−Re⁡tr⁡ ⁣((id⁡⊗TN)(PN) PN)]=0.\lim_{N\to\infty} \left[1-\operatorname{Re}\operatorname{tr}\!\left( (\operatorname{id}\otimes\mathcal T_N)(P_N)\,P_N \right)\right]=0.N→∞lim​[1−Retr((id⊗TN​)(PN​)PN​)]=0.

Then every subset J⊆{0,…,s−1}J\subseteq\{0,\ldots,s-1\}J⊆{0,…,s−1} satisfies

∑i∈JRi≥H(ρˉ)−H(ρˉJc),\sum_{i\in J}R_i\ge H(\bar\rho)-H(\bar\rho_{J^c}),i∈J∑​Ri​≥H(ρˉ​)−H(ρˉ​Jc​),

where ρˉJc\bar\rho_{J^c}ρˉ​Jc​ is obtained by retaining the coordinates in JcJ^cJc and summing over matching row and column indices in JJJ, and H(X)=−Re⁡tr⁡f(X)H(X)=-\operatorname{Re}\operatorname{tr}f(X)H(X)=−Retrf(X) for the functional-calculus function f(t)=tlog⁡2tf(t)=t\log_2tf(t)=tlog2​t, with f(0)=0f(0)=0f(0)=0. This includes the empty subset and the full subset; when s=0s=0s=0, the joint system is one-dimensional and the sole subset inequality is 0≥00\ge00≥0. All letters, including zero-probability letters, are subject to the purity and injectivity assumptions.

Human review
  • Endorsed by Shuze Chen · Oct 22, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Goku · Oct 22, 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