Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§6.1.2 (6.9)–(6.11): entanglement interpretations of min-entropy

Open
QAlgorithms.Papers.Independent.Intakenext100.P07.min_entropy_recovery

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​.

exp(−Hmin)=d_A max_E Tr[E_B→Aprime(ρ_AB)ψ_AAprime]. On normalized centers this is squared-fidelity recovery with lower bound −log d_A; the dual dimension-swapped formula (6.11) gives lower bound −log d_B.

2−Hmin⁡(A∣B)ρ=dAmax⁡E:B→A′Tr⁡((idA⊗E)(ρ)ψAA′)=dBmax⁡E:B′→ATr⁡(ρ(E⊗idB)(ψB′B)).2^{-H_{\min}(A|B)_\rho}=d_A\max_{\mathcal E:B\to A^\prime}\operatorname{Tr}((\mathrm{id}_A\otimes\mathcal E)(\rho)\psi_{AA^\prime})=d_B\max_{\mathcal E:B^\prime\to A}\operatorname{Tr}(\rho(\mathcal E\otimes\mathrm{id}_B)(\psi_{B^\prime B})).2−Hmin​(A∣B)ρ​=dA​E:B→A′max​Tr((idA​⊗E)(ρ)ψAA′​)=dB​E:B′→Amax​Tr(ρ(E⊗idB​)(ψB′B​)).

This result records the source’s operational or structural relation, with the stated endpoint and attainment clauses retained.

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.operationalmin
§6.1.2 (6.9)–(6.11): entanglement interpretations of min-entropy, arXiv:1504.00233v5, PDF p.93,94. -/
theorem min_entropy_recovery [Nonempty A] [Nonempty B] (r : paperMat (A × B)) (hr : paperSubnormalized r) :
    (∃ E : paperMat B →ₗ[ℂ] paperMat A, paperCPTP E ∧
      ENNReal.ofReal ((Fintype.card A:ℝ)*(idTensorSuperop E r*paperEntangled A).trace.re) = expE (-paperHmin r) ∧
      (∀ F : paperMat B →ₗ[ℂ] paperMat A, paperCPTP F →
        (idTensorSuperop F r*paperEntangled A).trace.re ≤ (idTensorSuperop E r*paperEntangled A).trace.re) ∧
      (IsDensityMatrix r → paperHmin r = -paperLogE ((Fintype.card A:ℝ)*paperFidelity (idTensorSuperop E r) (paperEntangled A)))) ∧
    (∃ E : paperMat B →ₗ[ℂ] paperMat A, paperCPTP E ∧
      ENNReal.ofReal ((Fintype.card B:ℝ)*(r*preparedEntangled E).trace.re) = expE (-paperHmin r) ∧
      (∀ F : paperMat B →ₗ[ℂ] paperMat A, paperCPTP F → (r*preparedEntangled F).trace.re ≤ (r*preparedEntangled E).trace.re)) ∧
    (IsDensityMatrix r → (-Real.logb 2 (Fintype.card A):EReal) ≤ paperHmin r ∧
      (-Real.logb 2 (Fintype.card B):EReal) ≤ paperHmin r) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P07

end
Source
Quantum Information Processing with Finite Resources -- Mathematical Foundations, arXiv:1504.00233v5, PDF p. 93, §6.1.2 (6.9)–(6.11): entanglement interpretations of min-entropy
Read-back

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

Let A,BA,BA,B be nonempty finite types with decidable equality, and let r⪰0r\succeq0r⪰0 on A×BA\times BA×B satisfy Re⁡tr⁡r≤1\operatorname{Re}\operatorname{tr}r\le1Retrr≤1, including r=0r=0r=0. Write X⪯YX\preceq YX⪯Y when Y−X⪰0Y-X\succeq0Y−X⪰0, and define

Hmin⁡(r)=sup⁡{ℓ∈R:∃s⪰0, Re⁡tr⁡s≤1, r⪯2−ℓ(1A⊗s)}.H_{\min}(r)= \sup\{\ell\in\mathbb R:\exists s\succeq0,\ \operatorname{Re}\operatorname{tr}s\le1,\ r\preceq2^{-\ell}(1_A\otimes s)\}.Hmin​(r)=sup{ℓ∈R:∃s⪰0, Retrs≤1, r⪯2−ℓ(1A​⊗s)}.

For a nonempty finite type TTT, let ΩT\Omega_TΩT​ be the matrix on T×TT\times TT×T with entries

(ΩT)(i,j),(i′,j′)={∣T∣−1i=j and i′=j′,0otherwise.(\Omega_T)_{(i,j),(i',j')}= \begin{cases} |T|^{-1}&i=j\ \text{and }i'=j',\\ 0&\text{otherwise}. \end{cases}(ΩT​)(i,j),(i′,j′)​={∣T∣−10​i=j and i′=j′,otherwise.​

For a complex-linear map EEE from matrices on BBB to matrices on AAA, let (idA⊗E)(r)(\mathrm{id}_A\otimes E)(r)(idA​⊗E)(r) denote blockwise application,

((idA⊗E)(r))(a,c),(a′,c′)=E ⁣((r(a,b),(a′,b′))b,b′)cc′,((\mathrm{id}_A\otimes E)(r))_{(a,c),(a',c')}= E\!\left((r_{(a,b),(a',b')})_{b,b'}\right)_{cc'},((idA​⊗E)(r))(a,c),(a′,c′)​=E((r(a,b),(a′,b′)​)b,b′​)cc′​,

and define a matrix TET_ETE​ on A×BA\times BA×B by

(TE)(a,b),(a′,b′)=((idB⊗E)(ΩB))(b,a),(b′,a′).(T_E)_{(a,b),(a',b')}= ((\mathrm{id}_B\otimes E)(\Omega_B))_{(b,a),(b',a')}.(TE​)(a,b),(a′,b′)​=((idB​⊗E)(ΩB​))(b,a),(b′,a′)​.

There exists a completely positive complex-linear map E:B→AE:B\to AE:B→A preserving the complex trace of every matrix such that

ofReal⁡ ⁣(∣A∣ Re⁡tr⁡((idA⊗E)(r) ΩA))=Exp⁡2(−Hmin⁡(r)),\operatorname{ofReal}\!\left( |A|\,\operatorname{Re}\operatorname{tr}((\mathrm{id}_A\otimes E)(r)\,\Omega_A) \right) =\operatorname{Exp}_2(-H_{\min}(r)),ofReal(∣A∣Retr((idA​⊗E)(r)ΩA​))=Exp2​(−Hmin​(r)),

and every other such map FFF satisfies

Re⁡tr⁡((idA⊗F)(r) ΩA)≤Re⁡tr⁡((idA⊗E)(r) ΩA).\operatorname{Re}\operatorname{tr}((\mathrm{id}_A\otimes F)(r)\,\Omega_A) \le \operatorname{Re}\operatorname{tr}((\mathrm{id}_A\otimes E)(r)\,\Omega_A).Retr((idA​⊗F)(r)ΩA​)≤Retr((idA​⊗E)(r)ΩA​).

For this same EEE, if rrr has complex trace 111, then

Hmin⁡(r)=−L ⁣(∣A∣(Re⁡tr⁡(idA⊗E)(r) ΩA (idA⊗E)(r))2).H_{\min}(r)= -\mathcal L\!\left( |A|\left(\operatorname{Re}\operatorname{tr} \sqrt{\sqrt{(\mathrm{id}_A\otimes E)(r)}\,\Omega_A\, \sqrt{(\mathrm{id}_A\otimes E)(r)}}\right)^2 \right).Hmin​(r)=−L(∣A∣(Retr(idA​⊗E)(r)​ΩA​(idA​⊗E)(r)​​)2).

There also exists a completely positive trace-preserving complex-linear map E:B→AE:B\to AE:B→A, chosen independently of the preceding existential witness, such that

ofReal⁡ ⁣(∣B∣ Re⁡tr⁡(rTE))=Exp⁡2(−Hmin⁡(r)),Re⁡tr⁡(rTF)≤Re⁡tr⁡(rTE)\operatorname{ofReal}\!\left(|B|\,\operatorname{Re}\operatorname{tr}(rT_E)\right) =\operatorname{Exp}_2(-H_{\min}(r)), \qquad \operatorname{Re}\operatorname{tr}(rT_F)\le \operatorname{Re}\operatorname{tr}(rT_E)ofReal(∣B∣Retr(rTE​))=Exp2​(−Hmin​(r)),Retr(rTF​)≤Retr(rTE​)

for every completely positive trace-preserving complex-linear F:B→AF:B\to AF:B→A. Finally, if rrr has trace 111, then

−log⁡2∣A∣≤Hmin⁡(r),−log⁡2∣B∣≤Hmin⁡(r)-\log_2|A|\le H_{\min}(r),\qquad -\log_2|B|\le H_{\min}(r)−log2​∣A∣≤Hmin​(r),−log2​∣B∣≤Hmin​(r)

in the extended reals. Complete positivity requires positivity of every finite amplification, including dimension 000; L(0)=−∞\mathcal L(0)=-\inftyL(0)=−∞ and otherwise L\mathcal LL embeds the real base-two logarithm; ofReal⁡\operatorname{ofReal}ofReal truncates below at 000; and Exp⁡2\operatorname{Exp}_2Exp2​ sends +∞+\infty+∞ to +∞+\infty+∞, −∞-\infty−∞ to 000, and finite hhh to ofReal⁡(2h)\operatorname{ofReal}(2^h)ofReal(2h).

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