Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem A.24: For finite-alphabet Xobs, compatible Yalg and Yalg-valued guessing Yobs on SAME alphabet, Pe=1−sum_j Tr(rho XjYj) gives H(Xobs|Yal

Open
QAlgorithms.Papers.Independent.Renew46.P43.fanoBound

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

prove2quantum

Let ρ\rhoρ be a normalized state, Xobs=(Xj)X_{\rm obs}=(X_j)Xobs​=(Xj​) a POVM on a finite nonempty alphabet XXX, and Y\mathcal YY an intrinsic subalgebra commuting with every XjX_jXj​. Let a guessing POVM (Yj)(Y_j)(Yj​) take values in Y\mathcal YY on the SAME alphabet. Write hhh for binary entropy in bits; a singleton alphabet has Pe=0P_e=0Pe​=0 and the continuous zero-product convention.

Pe=1−∑jTr⁡(ρXjYj),H(Xobs∣Y)≤h(Pe)+Pelog⁡2(∣X∣−1).P_e=1-\sum_j\operatorname{Tr}(\rho X_jY_j),\qquad H(X_{\rm obs}\mid\mathcal Y)\le h(P_e)+P_e\log_2(|X|-1).Pe​=1−j∑​Tr(ρXj​Yj​),H(Xobs​∣Y)≤h(Pe​)+Pe​log2​(∣X∣−1).

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 A.24, PDF p.75: same finite nonempty alphabet for the observable and
its guess; conditioning is on the whole compatible subsystem, not just its measurement. -/
theorem fanoBound {B : Type*} [Fintype B] [DecidableEq B] [Nonempty B] {Y A : IntrinsicSystem}
    (E : Embedding Y A) (X : B → Matrix (Fin A.dim) (Fin A.dim) ℂ)
    (hX : IsPOVM X) (hblocks : ∀ b, InBlocks A.block (X b))
    (hcompat : ∀ b T, InBlocks Y.block T → X b * E.map T = E.map T * X b)
    (guess : B → Matrix (Fin Y.dim) (Fin Y.dim) ℂ) (hguess : IsPOVM guess)
    (hguessblocks : ∀ b, InBlocks Y.block (guess b))
    (rho : Matrix (Fin A.dim) (Fin A.dim) ℂ) (hrho : A.IsState rho) :
    let Pe := 1 - ∑ b, effectProb rho (X b * E.map (guess b))
    vnEntropy (observableJoint E X rho) - viewedEntropy Y E.map rho ≤
      binEntropy2 Pe + Pe * Real.logb 2 (Fintype.card B - 1 : ℕ) := by sorry

end QAlgorithms.Papers.Independent.Renew46.P43

end
Source
Coding Theorems of Quantum Information Theory, arXiv:quant-ph/9907077v1, PDF p. 75, Theorem A.24
Read-back

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

Let BBB be a nonempty finite type with decidable equality, and let Y,AY,AY,A be systems with positive finite dimensions and natural-number block labels. Let EEE be a complex-linear map from matrices on YYY to matrices on AAA preserving the identity and block-diagonal matrices, multiplication and adjoints on block-diagonal matrices, and injective on those matrices; at every finite amplification, including dimension zero, it preserves positivity on matrices whose entries vanish between different YYY-block labels. Let (Xb)b∈B(X_b)_{b\in B}(Xb​)b∈B​ be positive semidefinite matrices on AAA summing to the identity, each zero between different AAA-block labels, and assume that XbE(T)=E(T)XbX_bE(T)=E(T)X_bXb​E(T)=E(T)Xb​ for every bbb and every block-diagonal matrix TTT on YYY. Let (Gb)b∈B(G_b)_{b\in B}(Gb​)b∈B​ be positive semidefinite matrices on YYY summing to the identity, each zero between different YYY-block labels. Let ρ\rhoρ be positive semidefinite of trace 111 on AAA, also zero between different AAA-block labels. Define

Pe=1−∑b∈BRe⁡tr⁡ ⁣(ρXbE(Gb)),Rij={tr⁡ ⁣(ρE(∣j⟩⟨i∣)),if i,j have the same Y-block label,0,otherwise,P_e=1-\sum_{b\in B}\operatorname{Re}\operatorname{tr}\!\left(\rho X_bE(G_b)\right), \qquad R_{ij}= \begin{cases} \operatorname{tr}\!\left(\rho E(|j\rangle\langle i|)\right), &\text{if }i,j\text{ have the same }Y\text{-block label},\\ 0,&\text{otherwise}, \end{cases}Pe​=1−b∈B∑​Retr(ρXb​E(Gb​)),Rij​={tr(ρE(∣j⟩⟨i∣)),0,​if i,j have the same Y-block label,otherwise,​

and define Ω\OmegaΩ by

Ω(b,i),(c,j)={tr⁡ ⁣(XbρXb E(∣j⟩⟨i∣)),b=c and i,j have the same Y-block label,0,otherwise.\Omega_{(b,i),(c,j)}= \begin{cases} \operatorname{tr}\!\left(\sqrt{X_b}\rho\sqrt{X_b}\,E(|j\rangle\langle i|)\right), &b=c\text{ and }i,j\text{ have the same }Y\text{-block label},\\ 0,&\text{otherwise}. \end{cases}Ω(b,i),(c,j)​={tr(Xb​​ρXb​​E(∣j⟩⟨i∣)),0,​b=c and i,j have the same Y-block label,otherwise.​

With H(M)=−Re⁡tr⁡[f(M)]H(M)=-\operatorname{Re}\operatorname{tr}[f(M)]H(M)=−Retr[f(M)], f(t)=tlog⁡2tf(t)=t\log_2tf(t)=tlog2​t applied by continuous functional calculus, and binary entropy h2(p)=[−pln⁡p−(1−p)ln⁡(1−p)]/ln⁡2h_2(p)=[-p\ln p-(1-p)\ln(1-p)]/\ln2h2​(p)=[−plnp−(1−p)ln(1−p)]/ln2, the conclusion is

H(Ω)−H(R)≤h2(Pe)+Pelog⁡2(∣B∣−1).H(\Omega)-H(R)\le h_2(P_e)+P_e\log_2(|B|-1).H(Ω)−H(R)≤h2​(Pe​)+Pe​log2​(∣B∣−1).

The logarithm uses the total real-logarithm convention log⁡20=0\log_2 0=0log2​0=0; thus the singleton case ∣B∣=1|B|=1∣B∣=1 is included and its logarithmic term is zero.

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