Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§6.3.1 (6.58): exact smoothing of flat entropy

Open
QAlgorithms.Papers.Independent.Intakenext100.P07.flat_entropy_smoothing

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

prove2quantum

All quantum systems below are finite-dimensional complex Hilbert spaces of positive dimension, and logarithms are binary. A normalized state is positive semidefinite with trace one; a subnormalized state has trace at most one. Marginals are partial traces. Here FFF is squared algebraic fidelity, F∗F_*F∗​ is squared generalized fidelity, P=1−F∗P=\sqrt{1-F_*}P=1−F∗​​ is purified distance, and Δ(ρ,τ)=max⁡0≤M≤I∣Tr⁡((ρ−τ)M)∣\Delta(\rho,\tau)=\max_{0\le M\le I}|\operatorname{Tr}((\rho-\tau)M)|Δ(ρ,τ)=max0≤M≤I​∣Tr((ρ−τ)M)∣. The symbols Hmin⁡H_{\min}Hmin​ and Hmax⁡H_{\max}Hmax​ denote conditional min- and max-entropy, with values +∞+\infty+∞ and −∞-\infty−∞ at zero. Smoothing optimizes over all subnormalized states in the purified-distance ball; every entropy radius satisfies 0≤ε<Tr⁡ρ0\le\varepsilon<\sqrt{\operatorname{Tr}\rho}0≤ε<Trρ​.

For a normalized state rho with equal min- and max-entropy and 0 <= epsilon < 1, both smoothing optima are attained by t=(1-epsilon^2)rho. In bits,

Hmin⁡ε(A∣B)ρ=Hmin⁡(A∣B)ρ−log⁡2(1−ε2),Hmax⁡ε(A∣B)ρ=Hmax⁡(A∣B)ρ+log⁡2(1−ε2).H_{\min}^{\varepsilon}(A|B)_\rho=H_{\min}(A|B)_\rho-\log_2(1-\varepsilon^2),\qquad H_{\max}^{\varepsilon}(A|B)_\rho=H_{\max}(A|B)_\rho+\log_2(1-\varepsilon^2).Hminε​(A∣B)ρ​=Hmin​(A∣B)ρ​−log2​(1−ε2),Hmaxε​(A∣B)ρ​=Hmax​(A∣B)ρ​+log2​(1−ε2).
Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P07
Formal statement
noncomputable section

open Filter

open scoped BigOperators Matrix ComplexConjugate ComplexOrder Matrix.Norms.L2Operator Topology

namespace QAlgorithms.Papers.Independent.Intakenext100.P07

attribute [local instance] Classical.propDecidable Classical.decEq

variable {I J A B C D X Y Z : Type*}

variable [Fintype I] [DecidableEq I] [Fintype J] [DecidableEq J]

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

variable [Fintype C] [DecidableEq C] [Fintype D] [DecidableEq D]

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

variable [Fintype Z] [DecidableEq Z]

/-- NODE paper:1504.00233#6.flatentropy
§6.3.1 (6.58) and its following max-entropy counterpart, PDF p.102.
Both optima are attained by the explicitly constructed (1-e^2)r. -/
theorem flat_entropy_smoothing [Nonempty A] [Nonempty B] (r : paperMat (A × B)) (hr : IsDensityMatrix r)
    (hf : paperHmin r = hmax r) (e : ℝ) (he : 0 ≤ e) (he1 : e < 1) :
    let t := ((1-e^2:ℝ):ℂ) • r
    t ∈ paperBall r e ∧ paperSmoothMin r e = paperHmin t ∧ paperSmoothMin r e = paperHmin r-(Real.logb 2 (1-e^2):EReal) ∧
    paperSmoothMax r e = hmax t ∧ paperSmoothMax r e = hmax r+(Real.logb 2 (1-e^2):EReal) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P07

end
Source
Quantum Information Processing with Finite Resources -- Mathematical Foundations, arXiv:1504.00233v5, PDF p. 102, §6.3.1 (6.58): exact smoothing of flat entropy
Read-back

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

Let AAA and BBB be finite, nonempty types with decidable equality, and let rrr be a complex matrix indexed by A×BA\times BA×B, positive semidefinite with Tr⁡r=1\operatorname{Tr}r=1Trr=1. To specify the quantities in the statement, call a matrix xxx subnormalized when x≥0x\ge0x≥0 and Re⁡Tr⁡x≤1\operatorname{Re}\operatorname{Tr}x\le1ReTrx≤1, and write

R(x,y)=Re⁡Tr⁡x y x,L(u)={−∞,u=0,log⁡2u,u≠0,R(x,y)=\operatorname{Re}\operatorname{Tr}\sqrt{\sqrt{x}\,y\,\sqrt{x}}, \qquad L(u)= \begin{cases} -\infty,&u=0,\\ \log_2u,&u\ne0, \end{cases}R(x,y)=ReTrx​yx​​,L(u)={−∞,log2​u,​u=0,u=0,​

where matrix square roots are defined by continuous functional calculus. Define the following extended-real quantities:

Hmin⁡(x)=sup⁡{ℓ∈R:∃σ≥0, Re⁡Tr⁡σ≤1, x≤2−ℓ(IA⊗σ)},Hmax⁡(x)=sup⁡σ≥0Re⁡Tr⁡σ≤1L ⁣(R(x,IA⊗σ)2),H_{\min}(x)= \sup\left\{\ell\in\mathbb R: \exists \sigma\ge0,\ \operatorname{Re}\operatorname{Tr}\sigma\le1,\ x\le 2^{-\ell}(I_A\otimes\sigma)\right\}, \qquad H_{\max}(x)= \sup_{\substack{\sigma\ge0\\\operatorname{Re}\operatorname{Tr}\sigma\le1}} L\!\left(R(x,I_A\otimes\sigma)^2\right),Hmin​(x)=sup{ℓ∈R:∃σ≥0, ReTrσ≤1, x≤2−ℓ(IA​⊗σ)},Hmax​(x)=σ≥0ReTrσ≤1​sup​L(R(x,IA​⊗σ)2),

where each σ\sigmaσ is indexed by BBB, and matrix inequalities mean that the difference is positive semidefinite. For a real number eee, define

Be(r)={x≥0:Re⁡Tr⁡x≤1, 1−(R(x,r)+(1−Re⁡Tr⁡x)(1−Re⁡Tr⁡r))2≤e},\mathcal B_e(r)= \left\{x\ge0: \operatorname{Re}\operatorname{Tr}x\le1,\ \sqrt{1-\left(R(x,r)+ \sqrt{(1-\operatorname{Re}\operatorname{Tr}x) (1-\operatorname{Re}\operatorname{Tr}r)}\right)^2}\le e \right\},Be​(r)={x≥0:ReTrx≤1, 1−(R(x,r)+(1−ReTrx)(1−ReTrr)​)2​≤e},

and define Hmin⁡e(r)=sup⁡x∈Be(r)Hmin⁡(x)H_{\min}^{e}(r)=\sup_{x\in\mathcal B_e(r)}H_{\min}(x)Hmine​(r)=supx∈Be​(r)​Hmin​(x) and Hmax⁡e(r)=inf⁡x∈Be(r)Hmax⁡(x)H_{\max}^{e}(r)=\inf_{x\in\mathcal B_e(r)}H_{\max}(x)Hmaxe​(r)=infx∈Be​(r)​Hmax​(x). The suprema and infimum defining these entropies are taken in the extended reals; the optimization matrices may be zero or have trace strictly below one. If Hmin⁡(r)=Hmax⁡(r)H_{\min}(r)=H_{\max}(r)Hmin​(r)=Hmax​(r), then for every real eee satisfying 0≤e<10\le e<10≤e<1, the matrix t=(1−e2)rt=(1-e^2)rt=(1−e2)r satisfies all five assertions

t∈Be(r),Hmin⁡e(r)=Hmin⁡(t),Hmin⁡e(r)=Hmin⁡(r)−log⁡2(1−e2),t\in\mathcal B_e(r),\qquad H_{\min}^{e}(r)=H_{\min}(t),\qquad H_{\min}^{e}(r)=H_{\min}(r)-\log_2(1-e^2),t∈Be​(r),Hmine​(r)=Hmin​(t),Hmine​(r)=Hmin​(r)−log2​(1−e2), Hmax⁡e(r)=Hmax⁡(t),Hmax⁡e(r)=Hmax⁡(r)+log⁡2(1−e2).H_{\max}^{e}(r)=H_{\max}(t),\qquad H_{\max}^{e}(r)=H_{\max}(r)+\log_2(1-e^2).Hmaxe​(r)=Hmax​(t),Hmaxe​(r)=Hmax​(r)+log2​(1−e2).

The real logarithms in these equalities are embedded into the extended reals. The permitted case e=0e=0e=0 gives t=rt=rt=r; throughout the stated range, 1−e2>01-e^2>01−e2>0.

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