Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Explicit block formula for ensemble alpha-Holevo information

Open
QAlgorithms.Papers.Independent.Support50.P19.cq_block_formula

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

prove2quantum

Let α>1\alpha>1α>1 and let {px,ρx}\{p_x,\rho_x\}{px​,ρx​} be a finite ensemble of density operators on a nonzero finite complex space, allowing zero probabilities. Define ρXQ=∑xpx∣x⟩⟨x∣⊗ρx\rho_{XQ}=\sum_xp_x|x\rangle\langle x|\otimes\rho_xρXQ​=∑x​px​∣x⟩⟨x∣⊗ρx​ and ρX=∑xpx∣x⟩⟨x∣\rho_X=\sum_xp_x|x\rangle\langle x|ρX​=∑x​px​∣x⟩⟨x∣. Then

χ~α({px,ρx}):=min⁡σD~α(ρXQ∥ρX⊗σ)=min⁡σ1α−1log⁡2∑xpxQ~α(ρx∥σ).\widetilde\chi_\alpha(\{p_x,\rho_x\}):=\min_\sigma\widetilde D_\alpha(\rho_{XQ}\Vert\rho_X\otimes\sigma)=\min_\sigma\frac1{\alpha-1}\log_2\sum_xp_x\widetilde Q_\alpha(\rho_x\Vert\sigma).χ​α​({px​,ρx​}):=σmin​Dα​(ρXQ​∥ρX​⊗σ)=σmin​α−11​log2​x∑​px​Q​α​(ρx​∥σ).

Both minima range over density operators and are attained. The alpha power is inside each trace defining Q~α\widetilde Q_\alphaQ​α​. A zero-weight unsupported block contributes zero. This formula computes the ensemble information from its conditional states.

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.17, (14): the cq definition equals the explicit support-aware block formula. -/
theorem cq_block_formula [Nonempty I] (a : ℝ) (ha : 1 < a) (e : Ensemble I) :
    renyiEnsembleInformation true a e.prob e.state = blockInformation a e.prob e.state ∧
    ∃ sigma : {X : Matrix I I ℂ // IsDensityMatrix X}, blockInformation a e.prob e.state =
      ((1 / (a - 1) : ℝ) : EReal) *
        elogNN (∑ x, ENNReal.ofReal (e.prob x) * sandwichedTrace a (e.state x) sigma.1) := 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. 17, eq14
Read-back

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

Let III be a finite nonempty index set with decidable equality, let a>1a>1a>1, and let eee be a finite ensemble consisting of a natural number mmm, nonnegative real weights pxp_xpx​ summing to one, and positive semidefinite matrices ρx\rho_xρx​ indexed by III, each with trace one. Define the block matrix Ω\OmegaΩ by Ω(x,i),(y,j)=1x=ypx(ρx)ij\Omega_{(x,i),(y,j)}=\mathbf1_{x=y}p_x(\rho_x)_{ij}Ω(x,i),(y,j)​=1x=y​px​(ρx​)ij​, and for a density matrix σ\sigmaσ define Ωσ\Omega_\sigmaΩσ​ by replacing every ρx\rho_xρx​ with σ\sigmaσ. For positive semidefinite X,YX,YX,Y, let qa(X,Y)=Re⁡tr⁡[(Y[(1−a)/(2a)]XY[(1−a)/(2a)])[a]]q_a(X,Y)=\operatorname{Re}\operatorname{tr}[(Y^{[(1-a)/(2a)]}XY^{[(1-a)/(2a)]})^{[a]}]qa​(X,Y)=Retr[(Y[(1−a)/(2a)]XY[(1−a)/(2a)])[a]], where spectral powers send zero eigenvalues to zero; set Qa(X,Y)=+∞Q_a(X,Y)=+\inftyQa​(X,Y)=+∞ under ker⁡Y⊈ker⁡X\ker Y\not\subseteq\ker XkerY⊆kerX, and otherwise Qa(X,Y)=max⁡{0,qa(X,Y)}Q_a(X,Y)=\max\{0,q_a(X,Y)\}Qa​(X,Y)=max{0,qa​(X,Y)}. Likewise set Da(X∥Y)=+∞D_a(X\|Y)=+\inftyDa​(X∥Y)=+∞ under support failure and otherwise Da(X∥Y)=L(qa(X,Y))/(a−1)D_a(X\|Y)=\mathcal L(q_a(X,Y))/(a-1)Da​(X∥Y)=L(qa​(X,Y))/(a−1), with L(t)=log⁡2t\mathcal L(t)=\log_2tL(t)=log2​t for t>0t>0t>0 and −∞-\infty−∞ otherwise. Writing L+(0)=−∞\mathcal L_+(0)=-\inftyL+​(0)=−∞, L+(+∞)=+∞\mathcal L_+(+\infty)=+\inftyL+​(+∞)=+∞, and L+(t)=log⁡2t\mathcal L_+(t)=\log_2tL+​(t)=log2​t for finite t>0t>0t>0, the assertion is

inf⁡σ≥0, tr⁡σ=1Da(Ω∥Ωσ)=inf⁡σ≥0, tr⁡σ=11a−1L+ ⁣(∑xpxQa(ρx,σ)),\inf_{\sigma\ge0,\ \operatorname{tr}\sigma=1}D_a(\Omega\|\Omega_\sigma) = \inf_{\sigma\ge0,\ \operatorname{tr}\sigma=1} \frac{1}{a-1}\mathcal L_+\!\left(\sum_xp_xQ_a(\rho_x,\sigma)\right),σ≥0, trσ=1inf​Da​(Ω∥Ωσ​)=σ≥0, trσ=1inf​a−11​L+​(x∑​px​Qa​(ρx​,σ)),

and there exists a density matrix σ\sigmaσ attaining the infimum on the right. The sum inside L+\mathcal L_+L+​ is in [0,+∞][0,+\infty][0,+∞], so a zero weight times +∞+\infty+∞ is zero. Zero weights and singular states are allowed; the condition ∑xpx=1\sum_xp_x=1∑x​px​=1 excludes m=0m=0m=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