Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Pure inputs attain all the output-purity extrema

Proved
QAlgorithms.Papers.Independent.Support50.P31.pureInputExtrema

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

prove2quantum

Let Φ\PhiΦ be a completely positive trace-preserving linear endomorphism of the operators on a nonzero finite-dimensional complex Hilbert space. A density matrix is a positive semidefinite operator of trace one. Write H(ρ)=−Tr⁡(ρlog⁡ρ)H(\rho)=-\operatorname{Tr}(\rho\log\rho)H(ρ)=−Tr(ρlogρ) in natural units, with 0log⁡0=00\log0=00log0=0; write ∥ρ∥p\|\rho\|_p∥ρ∥p​ for the Schatten ppp-norm, allowing p=∞p=\inftyp=∞ for the operator norm, and λmin⁡(ρ)\lambda_{\min}(\rho)λmin​(ρ) for the least eigenvalue including zero. There exist unit vectors attaining each of the following extrema:

max⁡ρ∥Φ(ρ)∥p=∥Φ(∣vp⟩⟨vp∣)∥p(1≤p≤∞),min⁡ρH(Φ(ρ))=H(Φ(∣vH⟩⟨vH∣)),min⁡ρλmin⁡(Φ(ρ))=λmin⁡(Φ(∣v−⟩⟨v−∣)).\begin{aligned} \max_\rho\|\Phi(\rho)\|_p &=\|\Phi(|v_p\rangle\langle v_p|)\|_p &&(1\le p\le\infty),\\ \min_\rho H(\Phi(\rho)) &=H(\Phi(|v_H\rangle\langle v_H|)),\\ \min_\rho\lambda_{\min}(\Phi(\rho)) &=\lambda_{\min}(\Phi(|v_-\rangle\langle v_-|)). \end{aligned}ρmax​∥Φ(ρ)∥p​ρmin​H(Φ(ρ))ρmin​λmin​(Φ(ρ))​=∥Φ(∣vp​⟩⟨vp​∣)∥p​=H(Φ(∣vH​⟩⟨vH​∣)),=λmin​(Φ(∣v−​⟩⟨v−​∣)).​(1≤p≤∞),

The extrema range over all density inputs. Different objectives may have different optimizing vectors. The assertion supplies pure optimizers and does not claim that every optimizer is pure.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P31
Formal statement
noncomputable section

open Filter Asymptotics

open scoped ComplexConjugate ENNReal

namespace QAlgorithms.Papers.Independent.Support50.P31

attribute [local instance] Classical.propDecidable Classical.decEq

open QAlgorithms.Papers.Independent.Information100.P56 (entropy)

open QAlgorithms.Papers.Independent.Information100.P63 (schatten)

open QAlgorithms.Papers.Independent.Gap48.P03 (holevoQuantity)

open QAlgorithms.Papers.Independent.Claimed50.P38 (tensorSuperop)

open QAlgorithms.Papers.Independent.Dependency50.P44 (tensorChannels)

/-- PDF pp.3–4: existence of pure optimizers for (4)–(6), not exclusivity.
Includes p = 1, infinity, and zero least eigenvalues. -/
theorem pureInputExtrema {I : Type} [Fintype I] [DecidableEq I] [Nonempty I]
    (Phi : Superop I I) (hPhi : IsChannel Phi) :
    (∀ p : ℝ≥0∞, 1 ≤ p → ∃ v : EuclideanSpace ℂ I, ‖v‖ = 1 ∧
      pPurity p (Phi (pureDensity v)) = nuP p Phi) ∧
    (∃ v : EuclideanSpace ℂ I, ‖v‖ = 1 ∧ entropy (Phi (pureDensity v)) = nuH Phi) ∧
    (∃ v : EuclideanSpace ℂ I, ‖v‖ = 1 ∧
      minEigenvalue (Phi (pureDensity v)) = nuMinus Phi) := by sorry

end QAlgorithms.Papers.Independent.Support50.P31

end
Source
On some additivity problems in quantum information theory, arXiv:math-ph/0003002v2, PDF p. 3, §2, pure-input attainment of output-purity extrema
Read-back

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

Let III be a finite nonempty type with decidable equality, and let Φ\PhiΦ map complex I×II\times II×I matrices to complex I×II\times II×I matrices. Assume that there exist a natural number mmm and matrices A0,…,Am−1A_0,\ldots,A_{m-1}A0​,…,Am−1​ such that ∑jAj†Aj=I\sum_jA_j^\dagger A_j=I∑j​Aj†​Aj​=I and Φ(X)=∑jAjXAj†\Phi(X)=\sum_jA_jXA_j^\daggerΦ(X)=∑j​Aj​XAj†​ for every matrix XXX. Write SI\mathcal S_ISI​ for the positive semidefinite matrices of trace one, and, for a density matrix XXX, define

Pp(X)={∥X∥op,p=∞,(∑jλj(X)r)1/r,p<∞, r=p,νp(Φ)=sup⁡ρ∈SIPp(Φ(ρ)),P_p(X)= \begin{cases} \|X\|_{\mathrm{op}},&p=\infty,\\ \left(\sum_j\lambda_j(X)^r\right)^{1/r},&p<\infty,\ r=p, \end{cases} \quad \nu_p(\Phi)=\sup_{\rho\in\mathcal S_I}P_p(\Phi(\rho)),Pp​(X)=⎩⎨⎧​∥X∥op​,(∑j​λj​(X)r)1/r,​p=∞,p<∞, r=p,​νp​(Φ)=ρ∈SI​sup​Pp​(Φ(ρ)),

where the eigenvalues are counted with multiplicity, and put H(X)=−∑jλj(X)log⁡λj(X)H(X)=-\sum_j\lambda_j(X)\log\lambda_j(X)H(X)=−∑j​λj​(X)logλj​(X) with natural logarithms and 0log⁡0=00\log0=00log0=0, νH(Φ)=inf⁡ρ∈SIH(Φ(ρ))\nu_H(\Phi)=\inf_{\rho\in\mathcal S_I}H(\Phi(\rho))νH​(Φ)=infρ∈SI​​H(Φ(ρ)), and ν−(Φ)=inf⁡ρ∈SIλmin⁡(Φ(ρ))\nu_-(\Phi)=\inf_{\rho\in\mathcal S_I}\lambda_{\min}(\Phi(\rho))ν−​(Φ)=infρ∈SI​​λmin​(Φ(ρ)). Then for every extended nonnegative real p≥1p\ge1p≥1, including p=∞p=\inftyp=∞, there exists a vector v∈CIv\in\mathbb C^Iv∈CI of Euclidean norm one such that Pp(Φ(vv†))=νp(Φ)P_p(\Phi(vv^\dagger))=\nu_p(\Phi)Pp​(Φ(vv†))=νp​(Φ); there also exists a unit vector attaining νH(Φ)\nu_H(\Phi)νH​(Φ), and there exists a unit vector attaining ν−(Φ)\nu_-(\Phi)ν−​(Φ). These three existential assertions do not require their vectors to be the same, and the first vector may depend on ppp.

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