Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma III.2: corner points lie in the rate region

Open
QAlgorithms.Papers.Independent.Support50.P09.corners_inRegion

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

prove2quantum

Let m≥1m\ge1m≥1 and let ∣ϕ⟩A1⋯AmR|\phi\rangle^{A_1\cdots A_mR}∣ϕ⟩A1​⋯Am​R be a normalized pure state on finite nonzero Hilbert spaces. All entropies are von Neumann entropies in bits of marginals of this same state. For K⊆[m]K\subseteq[m]K⊆[m], write AK=⨂i∈KAiA_K=\bigotimes_{i\in K}A_iAK​=⨂i∈K​Ai​ and

Cϕ(K)=12(∑i∈KH(Ai)+H(R)−H(AKR)).C_\phi(K)=\frac12\left(\sum_{i\in K}H(A_i)+H(R)-H(A_KR)\right).Cϕ​(K)=21​(i∈K∑​H(Ai​)+H(R)−H(AK​R)).

Let Rϕ\mathcal R_\phiRϕ​ be the closure of achievable nonnegative rate tuples: for every ϵ>0\epsilon>0ϵ>0, every sufficiently large positive blocklength admits independent local CPTP encoders and a joint receiver CPTP decoder, with log⁡2dim⁡Ci≤nQi\log_2\dim C_i\le nQ_ilog2​dimCi​≤nQi​ and squared fidelity at least 1−ϵ1-\epsilon1−ϵ with the entire original pure state on AnRnA^nR^nAnRn. The reference is untouched; no auxiliary communication or supplied shared entanglement is allowed.

Define PH={Q∈Rm:∑i∈KQi≥Cϕ(K) for every K⊆[m]}P_H=\{Q\in\mathbb R^m:\sum_{i\in K}Q_i\ge C_\phi(K)\text{ for every }K\subseteq[m]\}PH​={Q∈Rm:∑i∈K​Qi​≥Cϕ​(K) for every K⊆[m]}. For a permutation π\piπ, let Sπ,lS_{\pi,l}Sπ,l​ be its last lll participants and define the corner qπq^\piqπ by qπ(j)π=12I(Aπ(j);Aπ(j+1)⋯Aπ(m)R)q^\pi_{\pi(j)}=\tfrac12 I(A_{\pi(j)};A_{\pi(j+1)}\cdots A_{\pi(m)}R)qπ(j)π​=21​I(Aπ(j)​;Aπ(j+1)​⋯Aπ(m)​R). Write Q={qπ:π∈Sm}\mathcal Q=\{q^\pi:\pi\in S_m\}Q={qπ:π∈Sm​}; different permutations may give the same point.

All computed corners belong to the closed operational region:

Q⊆Rϕ.\mathcal Q\subseteq\mathcal R_\phi.Q⊆Rϕ​.

These rates supply the vertices used for time sharing and resource wasting.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P09
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Support50.P09

open QAlgorithms.Papers.Independent.Claimed50.P33

open scoped Matrix ComplexConjugate ComplexOrder

attribute [local instance] Classical.propDecidable Classical.decEq

section Compression

variable {m : ℕ} {D : Fin m → Type} [∀ i, Fintype (D i)] [hD : ∀ i, Nonempty (D i)]

variable {R : Type} [Fintype R] [hR : Nonempty R]

variable (phi : EuclideanSpace ℂ (RegisterBasis D × R)) (hphi : ‖phi‖ = 1)

include hphi hD hR

/-- Lemma III.2, PDF pp.7–8: computed suffix-based corners are in the closed region. -/
theorem corners_inRegion (hm : 0 < m) :
    compressionCorners (pureDensity phi) ⊆ multipartyCompressionRegion (pureDensity phi) := by sorry

end Compression

end QAlgorithms.Papers.Independent.Support50.P09

end
Source
Distributed Compression and Multiparty Squashed Entanglement, arXiv:0707.2792v2, PDF p. 7, Lemma III.2: corner points lie in the rate region
Read-back

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

For m>0m>0m>0, the ambient finite bases Di,RD_i,RDi​,R, and the ambient vector ϕ\phiϕ, put ρxy=ϕxϕy‾\rho_{xy}=\phi_x\overline{\phi_y}ρxy​=ϕx​ϕy​​. Write ρKR\rho_{KR}ρKR​ for the marginal retaining KKK and RRR, ρi\rho_iρi​ for the marginal of register iii alone, and S(X)=−Re⁡Tr⁡[f(X)]S(X)=-\operatorname{Re}\operatorname{Tr}[f(X)]S(X)=−ReTr[f(X)], with f(x)=xlog⁡2xf(x)=x\log_2xf(x)=xlog2​x, f(0)=0f(0)=0f(0)=0, applied by continuous functional calculus. For each permutation π\piπ of {0,…,m−1}\{0,\ldots,m-1\}{0,…,m−1}, let Li={π(j):π−1(i)<j}L_i=\{\pi(j):\pi^{-1}(i)<j\}Li​={π(j):π−1(i)<j} and define the real vector

viπ=S(ρi)+S(ρLiR)−S(ρ(Li∪{i})R)2.v^\pi_i=\frac{S(\rho_i)+S(\rho_{L_iR})-S(\rho_{(L_i\cup\{i\})R})}{2}.viπ​=2S(ρi​)+S(ρLi​R​)−S(ρ(Li​∪{i})R​)​.

Every such vector belongs to the closure of the nonnegative rate vectors QQQ for which, for every ε>0\varepsilon>0ε>0, all sufficiently large positive integers sss admit positive dimensions cic_ici​, independent encoders Ei\mathcal E_iEi​, and a joint decoder D\mathcal DD, satisfying log⁡2ci≤sQi\log_2c_i\le sQ_ilog2​ci​≤sQi​ and Re⁡Tr⁡[Ωs(id⁡Rs⊗(D∘⨂iEi))(Ωs)]≥1−ε\operatorname{Re}\operatorname{Tr}[\Omega_s(\operatorname{id}_{R^s}\otimes(\mathcal D\circ\bigotimes_i\mathcal E_i))(\Omega_s)]\ge1-\varepsilonReTr[Ωs​(idRs​⊗(D∘⨂i​Ei​))(Ωs​)]≥1−ε. Here Ωs=ρ⊗s\Omega_s=\rho^{\otimes s}Ωs​=ρ⊗s is reordered as Rs×∏iDisR^s\times\prod_iD_i^sRs×∏i​Dis​, and each map has a finite Kraus representation with ∑jEj†Ej=I\sum_jE_j^\dagger E_j=I∑j​Ej†​Ej​=I; each encoder outputs a cic_ici​-dimensional system and the decoder returns all sender registers. No norm-one hypothesis on ϕ\phiϕ appears in the displayed declaration.

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