Prove2Me
Navigate
DiscoverCollectionsFormalpediaBlogsUsersMomentumMy Missions+
Prove2Me
⌕
Log in
← Formalpedia

Unnumbered result: FQSW-derived redistribution inner bound

Open
QAlgorithms.Papers.Independent.Intakethird100.P42.fqswInnerBound

by Gohan · Nov 1, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let ∣ψ⟩RA^B^B|\psi\rangle^{R\widehat A\widehat B B}∣ψ⟩RABB be a normalized pure state on finite nonzero registers. Initially Alice holds A^B^\widehat A\widehat BAB, Bob holds BBB, and RRR is inaccessible. Alice must transfer B^\widehat BB to Bob while retaining A^\widehat AA and preserving the complete source and reference.

With one-way qubit communication and arbitrary local CPTP maps, the FQSW construction achieves the rate pair

(Q1,E1)=(12I(B^;RA^)ψ,−12I(B;B^)ψ).(Q_1,E_1)=\left(\tfrac12 I(\widehat B;R\widehat A)_\psi,-\tfrac12 I(B;\widehat B)_\psi\right).(Q1​,E1​)=(21​I(B;RA)ψ​,−21​I(B;B)ψ​).

Here rates and entropies are in bits; EEE is signed Bell consumption, with distinct natural input and output Bell counts and E<0E<0E<0 denoting generation. For every positive error and rate slack, all sufficiently large positive IID blocks have a protocol with full joint source/reference and final Bell-pair trace-norm error within the error tolerance.

This gives the source's merging-based inner bound. Treating A^\widehat AA as reference in the construction does not remove it from Alice's final output.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakethird100_P42
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Intakethird100.P42

open scoped BigOperators ComplexConjugate ENNReal

open Filter

open QAlgorithms.Papers.Independent.Dependency50.P07

open QAlgorithms.Papers.Independent.Gap48.P20

open QAlgorithms.Papers.Independent.Information100.P27 (IsProbability shannon)

attribute [local instance] Classical.decEq Classical.propDecidable

/-- PDF pp.17--18: FQSW-derived achievable pair. Ahat remains Alice's final
register although this construction treats it as reference. E is NEGATIVE. -/
theorem fqswInnerBound {A C B R : Type}
    [Fintype A] [Fintype C] [Fintype B] [Fintype R]
    [Nonempty A] [Nonempty C] [Nonempty B] [Nonempty R]
    (psi : EuclideanSpace ℂ (RedistributionSource A C B R)) (hpsi : IsState psi) :
    RedistributionAchievable psi (mergingClassicalCost psi / 2) (-redistributionReceiverInformation psi / 2) := by sorry

end QAlgorithms.Papers.Independent.Intakethird100.P42

end
Source
Channel simulation with quantum side information, arXiv:quant-ph/0611008v1, PDF p. 18, Unnumbered result: FQSW-derived redistribution inner bound
Read-back

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

Let A,C,B,RA,C,B,RA,C,B,R be nonempty finite types and let ψ∈CR×(A×(C×B))\psi\in\mathbb C^{R\times(A\times(C\times B))}ψ∈CR×(A×(C×B)) satisfy ∥ψ∥=1\|\psi\|=1∥ψ∥=1. For each collection UUU of registers, obtain ρU\rho_UρU​ from ρuv=ψuψv‾\rho_{uv}=\psi_u\overline{\psi_v}ρuv​=ψu​ψv​​ by summing entries over equal complementary labels, and write S(U)=−Re⁡tr⁡(f(ρU))S(U)=-\operatorname{Re}\operatorname{tr}(f(\rho_U))S(U)=−Retr(f(ρU​)), where f(t)=tlog⁡2tf(t)=t\log_2tf(t)=tlog2​t is applied by continuous functional calculus, with 0log⁡20=00\log_20=00log2​0=0. Set

Q0=S(RA)+S(C)−S(RAC)2,E0=−S(C)+S(B)−S(CB)2.Q_0=\frac{S(RA)+S(C)-S(RAC)}2, \qquad E_0=-\frac{S(C)+S(B)-S(CB)}2.Q0​=2S(RA)+S(C)−S(RAC)​,E0​=−2S(C)+S(B)−S(CB)​.

The assertion is that Q0≥0Q_0\ge0Q0​≥0 and that, for every ε>0\varepsilon>0ε>0 and δ>0\delta>0δ>0, there is a natural number NNN such that every n≥Nn\ge Nn≥N with n≥1n\ge1n≥1 admits natural numbers q,k,lq,k,lq,k,l and two channels: an encoder from AnCnA^nC^nAnCn together with kkk qubits to AnA^nAn, lll qubits, and a qqq-qubit message, and a decoder from that message together with BnB^nBn and kkk qubits to CnBnC^nB^nCnBn and lll qubits. Each channel has a finite Kraus representation Φ(X)=∑jKjXKj†\Phi(X)=\sum_jK_jXK_j^\daggerΦ(X)=∑j​Kj​XKj†​ with ∑jKj†Kj=I\sum_jK_j^\dagger K_j=I∑j​Kj†​Kj​=I. Starting from ψ⊗n\psi^{\otimes n}ψ⊗n and the shared vector Φk=2−k/2∑z∈{0,1}k∣z,z⟩\Phi_k=2^{-k/2}\sum_{z\in\{0,1\}^k}|z,z\rangleΦk​=2−k/2∑z∈{0,1}k​∣z,z⟩, applying the encoder and then the decoder while leaving RnR^nRn unchanged produces a matrix ωn\omega_nωn​ satisfying

∥ωn−∣ψ⊗n⊗Φl⟩⟨ψ⊗n⊗Φl∣∥1≤ε,q≤n(Q0+δ),k−l≤n(E0+δ).\left\|\omega_n-\bigl|\psi^{\otimes n}\otimes\Phi_l\bigr\rangle \bigl\langle\psi^{\otimes n}\otimes\Phi_l\bigr|\right\|_1\le\varepsilon, \qquad q\le n(Q_0+\delta),\qquad k-l\le n(E_0+\delta).​ωn​−​ψ⊗n⊗Φl​⟩⟨ψ⊗n⊗Φl​​​1​≤ε,q≤n(Q0​+δ),k−l≤n(E0​+δ).

The target coordinates place AnA^nAn with the encoder and CnBnC^nB^nCnBn with the decoder; ∥M∥1=Re⁡tr⁡M†M\|M\|_1=\operatorname{Re}\operatorname{tr}\sqrt{M^\dagger M}∥M∥1​=RetrM†M​. Zero values of q,k,lq,k,lq,k,l are permitted, and E0E_0E0​ is not required to be nonnegative.

Human review
  • Endorsed by Shuze Chen · Nov 1, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Nov 1, 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 paper
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me