Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Weak converse with binary entropy

Open
QAlgorithms.Papers.Independent.Support50.P19.weakConverse

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

prove2quantum

Let N:A→B\mathcal N:A\to BN:A→B be a finite-dimensional channel, n≥1n\ge1n≥1 and M≥1M\ge1M≥1 integers, and let a uniform-message unassisted block code have average success psuccp_{\rm succ}psucc​ and actual rate R=log⁡2M/nR=\log_2M/nR=log2​M/n. If 1−psucc≤ε1-p_{\rm succ}\le\varepsilon1−psucc​≤ε, 0≤ε<10\le\varepsilon<10≤ε<1, and ε≤1−1/M\varepsilon\le1-1/Mε≤1−1/M, then

R≤χ(N⊗n)+h2(ε)n(1−ε),h2(t)=−tlog⁡2t−(1−t)log⁡2(1−t).R\le\frac{\chi(\mathcal N^{\otimes n})+h_2(\varepsilon)}{n(1-\varepsilon)},\qquad h_2(t)=-t\log_2t-(1-t)\log_2(1-t).R≤n(1−ε)χ(N⊗n)+h2​(ε)​,h2​(t)=−tlog2​t−(1−t)log2​(1−t).

Here χ\chiχ is the single-letter Holevo information of the entire block channel, and 0log⁡20=00\log_2 0=00log2​0=0. The bound applies to arbitrary mixed block codewords and joint POVMs, and gives the usual weak converse.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P19
Formal statement
noncomputable section

open scoped Matrix ComplexOrder ComplexConjugate ENNReal

open Filter Topology

namespace QAlgorithms.Papers.Independent.Support50.P19

attribute [local instance] Classical.propDecidable Classical.decEq

open QAlgorithms.Papers.Independent.Dependency50.P35
  (tr Dominates spectralPower elog elogNN rawTrace cq)

open QAlgorithms.Papers.Independent.Dependency50.P09 (sandwichedD petzD relativeD)

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

open QAlgorithms.Papers.Independent.Information100.P27 (average)

open QAlgorithms.Papers.Independent.Information100.P67
  (IsIsometry receiverChannel complementaryChannel)

open QAlgorithms.Papers.Independent.Claimed50.P48 (FiniteKrausCP)

open QAlgorithms.Papers.Independent.Dependency50.P23
  (UACode uaCodeSuccess ordinaryClassicalCapacity IsEB IsHadamard)

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

open QAlgorithms.Papers.Independent.Information100.P71 (channelPower)

open QAlgorithms.Papers.Independent.Information100.P62 (tensorChannel)

open QAlgorithms.Papers.Independent.Information100.P63 (tensorPower)

open QAlgorithms.Papers.Independent.Claimed50.P26
  (CovariantIrreducible minOutput UnassistedExponentialConverse)

variable {I J K L : Type} [Fintype I] [Fintype J] [Fintype K] [Fintype L]

variable [DecidableEq I] [DecidableEq J] [DecidableEq K] [DecidableEq L]

/-- PDF p.11, Remark 3: weak converse for an allowed error bound, with the p.10 convention. -/
theorem weakConverse [Nonempty I] [Nonempty J] (N : Superop I J) (hN : IsChannel N)
    (n M : ℕ) (hn : 0 < n) (hM : 0 < M)
    (C : UACode (Fin n → I) (Fin n → J) M)
    (e : ℝ) (he : 0 ≤ e) (he1 : e < 1) (heM : e ≤ 1 - 1 / (M : ℝ))
    (herr : 1 - uaCodeSuccess (channelPower N n) C ≤ e) :
    (Real.logb 2 (M : ℝ) / (n : ℝ)) ≤ ((QAlgorithms.Papers.Independent.Claimed50.P26.channelChi 1 (channelPower N n)) + binEntropy2 e) / ((n : ℝ) * (1 - e)) := by sorry

end QAlgorithms.Papers.Independent.Support50.P19

end
Source
Strong converse for the classical capacity of entanglement-breaking and Hadamard channels via a sandwiched Renyi relative entropy, arXiv:1306.1586v4, PDF p. 11, remark3
Read-back

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

Let I,JI,JI,J be finite nonempty types with decidable equality, and let NNN be a channel, so N(X)=∑sEsXEs†N(X)=\sum_sE_sXE_s^\daggerN(X)=∑s​Es​XEs†​ for a finite family satisfying ∑sEs†Es=1\sum_sE_s^\dagger E_s=1∑s​Es†​Es​=1. Let n,Mn,Mn,M be positive natural numbers, and let CCC consist of density matrices ρm\rho_mρm​ on InI^nIn and positive semidefinite measurement matrices DmD_mDm​ on JnJ^nJn, indexed by 0≤m<M0\le m<M0≤m<M, with ∑mDm=1\sum_mD_m=1∑m​Dm​=1. Define

N[n](X)y,z=∑x,x′∈InXx,x′∏i=0n−1N(Exi,xi′)yi,zi,PC=1M∑m=0M−1Re⁡Tr⁡ ⁣(DmN[n](ρm)),N^{[n]}(X)_{y,z} =\sum_{x,x'\in I^n}X_{x,x'} \prod_{i=0}^{n-1}N(E_{x_i,x'_i})_{y_i,z_i}, \qquad P_C=\frac1M\sum_{m=0}^{M-1} \operatorname{Re}\operatorname{Tr}\!\left(D_mN^{[n]}(\rho_m)\right),N[n](X)y,z​=x,x′∈In∑​Xx,x′​i=0∏n−1​N(Exi​,xi′​​)yi​,zi​​,PC​=M1​m=0∑M−1​ReTr(Dm​N[n](ρm​)),

where Eu,vE_{u,v}Eu,v​ denotes a matrix unit. For every real eee satisfying 0≤e<10\le e<10≤e<1, e≤1−1/Me\le1-1/Me≤1−1/M, and 1−PC≤e1-P_C\le e1−PC​≤e, the theorem asserts

log⁡2Mn≤χ1(N[n])+h2(e)n(1−e).\frac{\log_2M}{n} \le \frac{\chi_1(N^{[n]})+h_2(e)}{n(1-e)}.nlog2​M​≤n(1−e)χ1​(N[n])+h2​(e)​.

Here χ1(T)\chi_1(T)χ1​(T) is the real supremum, over all finite ensembles of density matrices (px,τx)(p_x,\tau_x)(px​,τx​) on the input space with px≥0p_x\ge0px​≥0 and ∑xpx=1\sum_xp_x=1∑x​px​=1, of

H ⁣(T ⁣(∑xpxτx))−∑xpxH(T(τx)),H(X)=−Re⁡Tr⁡cfc⁡[t↦tlog⁡2t](X).H\!\left(T\!\left(\sum_xp_x\tau_x\right)\right) -\sum_xp_xH(T(\tau_x)), \qquad H(X)=-\operatorname{Re}\operatorname{Tr} \operatorname{cfc}[t\mapsto t\log_2t](X).H(T(x∑​px​τx​))−x∑​px​H(T(τx​)),H(X)=−ReTrcfc[t↦tlog2​t](X).

The scalar h2(e)h_2(e)h2​(e) is [−eln⁡e−(1−e)ln⁡(1−e)]/ln⁡2\bigl[-e\ln e-(1-e)\ln(1-e)\bigr]/\ln2[−elne−(1−e)ln(1−e)]/ln2, with the total logarithm convention ln⁡0=0\ln0=0ln0=0. The assumptions make n(1−e)>0n(1-e)>0n(1−e)>0; the case M=1M=1M=1 is included and forces e=0e=0e=0.

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