Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Single-letter public rate for classical outputs

Open
QAlgorithms.Papers.Independent.Intakenext100.P30.classical_public_rate

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

prove2quantum

Suppose all receiver output states of the fixed cq channel Θ\ThetaΘ are simultaneously diagonalizable in one orthonormal basis. The operational public iterated limit exists and satisfies

Rpub(Θ)=sup⁡PX[H(X)−H(X∣Y)].R_{\rm pub}(\Theta)=\sup_{P_X}\bigl[H(X)-H(X|Y)\bigr].Rpub​(Θ)=PX​sup​[H(X)−H(X∣Y)].

The rate means lim⁡ε↓0lim⁡n→∞Cpubε(Θ⊗n)/n\lim_{\varepsilon\downarrow0}\lim_{n\to\infty}C_{\rm pub}^{\varepsilon}(\Theta^{\otimes n})/nlimε↓0​limn→∞​Cpubε​(Θ⊗n)/n, with existence asserted. Only reliability is required. This recovers the classical single-letter public formula.

All alphabets are finite and nonempty, and output Hilbert spaces are finite dimensional and nonzero. A classical–quantum channel assigns a positive semidefinite, trace-one matrix to each input. Encoders may use private randomness; decoders are normalized POVMs. Codes are fixed independently of the message, with no supplied shared key or entanglement. Logarithms and entropies are in bits.

The maximum is attained: there exists a normalized input law PXP_XPX​ whose displayed mutual information equals the operational rate.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P30
Formal statement
noncomputable section

open scoped BigOperators ComplexConjugate Matrix ComplexOrder

open Filter

namespace QAlgorithms.Papers.Independent.Intakenext100.P30

variable {X Y Z M U V A B C : Type*}

variable [Fintype X] [DecidableEq X] [Fintype Y] [DecidableEq Y]

variable [Fintype Z] [DecidableEq Z] [Fintype M] [DecidableEq M]

variable [Fintype U] [DecidableEq U] [Fintype V] [DecidableEq V]

variable [Fintype A] [DecidableEq A] [Fintype B] [DecidableEq B]

variable [Fintype C] [DecidableEq C]

attribute [local instance] Classical.propDecidable

/-- Section IV, PDF p.5, (18): classical-output single-letter public rate.
The receiver states have a common diagonalizing basis; secrecy is not imposed.
An optimizing input law is a conclusion, retaining the printed maximum. -/
theorem classical_public_rate [Nonempty X] [Nonempty Y]
    (theta : X → Matrix Y Y ℂ) (hTheta : RRIsCq theta) (hclassical : RRClassicalOutputs theta) :
    ∃ rate : ℝ, RRPublicRate theta rate ∧ (rate : EReal) = rrPublicInformation theta ∧
      ∃ p : RRDistribution X,
        rate = vnEntropy (Matrix.diagonal (fun x => (p.prob x : ℂ))) -
          rrConditionalEntropy (rrCqState p.prob theta) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P30

end
Source
Noisy channel coding via privacy amplification and information reconciliation, arXiv:1012.4814v2, PDF p. 5, §IV (18): classical single-letter public rate
Read-back

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

Let X,YX,YX,Y be nonempty finite types with decidable equality, and let θx\theta_xθx​, for x∈Xx\in Xx∈X, be positive semidefinite matrices on YYY with trace 111. Suppose there exists a matrix VVV with V†V=VV†=IYV^\dagger V=VV^\dagger=I_YV†V=VV†=IY​ such that every V†θxVV^\dagger\theta_xVV†θx​V has all off-diagonal entries zero. For a matrix AAA, write S(A)=−Re⁡tr⁡f(A)S(A)=-\operatorname{Re}\operatorname{tr}f(A)S(A)=−Retrf(A), where f(t)=tlog⁡2tf(t)=t\log_2tf(t)=tlog2​t and f(A)f(A)f(A) is defined by continuous functional calculus, using the total real logarithm with log⁡20=0\log_2 0=0log2​0=0. For a probability distribution ppp on XXX, define Θ(p)(x,y),(x′,y′)=1x=x′p(x)(θx)y,y′\Theta(p)_{(x,y),(x',y')}=\mathbf1_{x=x'}p(x)(\theta_x)_{y,y'}Θ(p)(x,y),(x′,y′)​=1x=x′​p(x)(θx​)y,y′​ and

J(p)=S(diag⁡p)−S(Θ(p))+S ⁣(∑xp(x)θx).J(p)=S(\operatorname{diag}p)-S(\Theta(p)) +S\!\left(\sum_xp(x)\theta_x\right).J(p)=S(diagp)−S(Θ(p))+S(x∑​p(x)θx​).

For each positive integer nnn and real ε\varepsilonε, define Cn(ε)C_n(\varepsilon)Cn​(ε) to be the extended-real supremum of log⁡2L\log_2Llog2​L over all integers L≥1L\ge1L≥1, all probability distributions eme_mem​ on XnX^nXn for m∈{1,…,L}m\in\{1,\ldots,L\}m∈{1,…,L}, and all positive semidefinite matrices DmD_mDm​ on YnY^nYn summing to IYnI_{Y^n}IYn​, subject to

1−Re⁡tr⁡ ⁣(Dm∑x∈Xnem(x)θx(n))≤εfor every m,(θx(n))a,b=∏i=1n(θxi)ai,bi.1-\operatorname{Re}\operatorname{tr}\!\left( D_m\sum_{x\in X^n}e_m(x)\theta^{(n)}_x\right)\le\varepsilon \quad\text{for every }m, \qquad (\theta^{(n)}_x)_{a,b}=\prod_{i=1}^n(\theta_{x_i})_{a_i,b_i}.1−Retr(Dm​x∈Xn∑​em​(x)θx(n)​)≤εfor every m,(θx(n)​)a,b​=i=1∏n​(θxi​​)ai​,bi​​.

Then there exists a real number rrr and a function ℓ:R→R‾\ell:\mathbb R\to\overline{\mathbb R}ℓ:R→R such that, for every 0<ε<1/20<\varepsilon<1/20<ε<1/2, Cn(ε)/nC_n(\varepsilon)/nCn​(ε)/n converges to ℓ(ε)\ell(\varepsilon)ℓ(ε) as n→∞n\to\inftyn→∞ through positive integers, and ℓ(ε)\ell(\varepsilon)ℓ(ε) converges to the embedded real number rrr as ε→0\varepsilon\to0ε→0 from above. Furthermore,

(r:R‾)=sup⁡p(J(p):R‾),r=J(p∗)(r:\overline{\mathbb R})=\sup_{p} (J(p):\overline{\mathbb R}), \qquad r=J(p_*)(r:R)=psup​(J(p):R),r=J(p∗​)

for some probability distribution p∗p_*p∗​ on XXX; the supremum ranges over all nonnegative ppp with ∑xp(x)=1\sum_xp(x)=1∑x​p(x)=1, including distributions with zero entries, and singleton XXX and YYY are allowed.

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