Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Quantitative comparison near order one

Open
QAlgorithms.Papers.Independent.Support50.P19.quantitative_relative

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

prove2quantum

Let ρ,σ\rho,\sigmaρ,σ be density operators on a nonzero finite complex space with supp⁡ρ⊆supp⁡σ\operatorname{supp}\rho\subseteq\operatorname{supp}\sigmasuppρ⊆suppσ. Define

ν=2D3/2(ρ∥σ)/2+2−D1/2(ρ∥σ)/2+1.\nu=2^{D_{3/2}(\rho\Vert\sigma)/2}+2^{-D_{1/2}(\rho\Vert\sigma)/2}+1.ν=2D3/2​(ρ∥σ)/2+2−D1/2​(ρ∥σ)/2+1.

For 1<α<1+log⁡23/(4log⁡2ν)1<\alpha<1+\log_2 3/(4\log_2\nu)1<α<1+log2​3/(4log2​ν),

D~α(ρ∥σ)≤D(ρ∥σ)+4(α−1)(log⁡2ν)2.\widetilde D_\alpha(\rho\Vert\sigma)\le D(\rho\Vert\sigma)+4(\alpha-1)(\log_2\nu)^2.Dα​(ρ∥σ)≤D(ρ∥σ)+4(α−1)(log2​ν)2.

Here DtD_tDt​ denotes Petz divergence in bits, and DDD ordinary relative entropy. The local definition computes the two powers of two by their equal finite trace expressions, so no infinite divergence is silently converted to a real zero. This estimate controls the limit quantitatively.

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.22, (21), with (20): strict alpha cap and the half factors in renyiNu. -/
theorem quantitative_relative [Nonempty I] (rho sigma : {X : Matrix I I ℂ // IsDensityMatrix X}) (hsupp : Dominates sigma.1 rho.1)
    (a : ℝ) (ha : 1 < a) (hcap : a < 1 + Real.logb 2 3 / (4 * Real.logb 2 (renyiNu rho.1 sigma.1))) :
    sandwichedD a rho.1 sigma.1 ≤ relativeD rho.1 sigma.1 +
      ((4 * (a - 1) * (Real.logb 2 (renyiNu rho.1 sigma.1)) ^ 2 : ℝ) : EReal) := 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. 22, eq21
Read-back

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

Let III be a finite nonempty index type with decidable equality, and let ρ,σ\rho,\sigmaρ,σ be positive semidefinite complex matrices on III, each of trace 111, satisfying ker⁡σ⊆ker⁡ρ\ker\sigma\subseteq\ker\rhokerσ⊆kerρ. Define

ν=Re⁡tr⁡(ρ[3/2]σ[−1/2])+Re⁡tr⁡(ρ[1/2]σ[1/2])+1,\nu=\operatorname{Re}\operatorname{tr}(\rho^{[3/2]}\sigma^{[-1/2]}) +\operatorname{Re}\operatorname{tr}(\rho^{[1/2]}\sigma^{[1/2]})+1,ν=Retr(ρ[3/2]σ[−1/2])+Retr(ρ[1/2]σ[1/2])+1,

where X[t]X^{[t]}X[t] applies the real power xtx^txt to each nonzero eigenvalue and sends zero eigenvalues to 000. For every real aaa satisfying

1<a<1+log⁡234log⁡2ν,1<a<1+\frac{\log_2 3}{4\log_2\nu},1<a<1+4log2​νlog2​3​,

the following extended-real inequality holds:

Da(ρ∥σ)≤D(ρ∥σ)+4(a−1)(log⁡2ν)2.D_a(\rho\|\sigma)\le D(\rho\|\sigma) +4(a-1)(\log_2\nu)^2.Da​(ρ∥σ)≤D(ρ∥σ)+4(a−1)(log2​ν)2.

Here D(ρ∥σ)=Re⁡tr⁡[ρlog⁡2ρ−ρlog⁡2σ]D(\rho\|\sigma)=\operatorname{Re}\operatorname{tr}[\rho\log_2\rho-\rho\log_2\sigma]D(ρ∥σ)=Retr[ρlog2​ρ−ρlog2​σ], with the matrix logarithms assigning value 000 at zero eigenvalues, and

Da(ρ∥σ)=1a−1log⁡2Re⁡tr⁡[(σ[(1−a)/(2a)]ρσ[(1−a)/(2a)])[a]].D_a(\rho\|\sigma)=\frac1{a-1}\log_2\operatorname{Re}\operatorname{tr} \left[\left(\sigma^{[(1-a)/(2a)]}\rho \sigma^{[(1-a)/(2a)]}\right)^{[a]}\right].Da​(ρ∥σ)=a−11​log2​Retr[(σ[(1−a)/(2a)]ρσ[(1−a)/(2a)])[a]].

The real correction term is embedded into the extended reals. In the underlying definitions both divergences are +∞+\infty+∞ upon failure of the stated kernel inclusion; that failure is excluded here. No upper bound a≤2a\le2a≤2 is a hypothesis of this declaration.

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