Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Section III.C: Schmidt phase randomization and the attained classical correlation

Open
QAlgorithms.Papers.Independent.Intakenext100.P47.schmidtPhase_disentangling

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

prove2quantum

Let A,BA,BA,B be finite nonzero complex spaces and let D≥1D\ge1D≥1. Let aj∈Aa_j\in Aaj​∈A and bj∈Bb_j\in Bbj​∈B be orthonormal families indexed by j=0,…,D−1j=0,\ldots,D-1j=0,…,D−1, and let fj>0f_j>0fj​>0, ∑jfj=1\sum_j f_j=1∑j​fj​=1. Put ∣Φ⟩=∑jfj∣ajbj⟩|\Phi\rangle=\sum_j\sqrt{f_j}|a_jb_j\rangle∣Φ⟩=∑j​fj​​∣aj​bj​⟩ and define Ukaj=e2πikj/DajU_k a_j=e^{2\pi\mathrm{i}kj/D}a_jUk​aj​=e2πikj/Daj​, extending UkU_kUk​ by the identity on the unused orthogonal complement. The DDD-term uniform Alice-only ensemble has output

1D∑k=0D−1(Uk⊗1)∣Φ⟩⟨Φ∣(Uk⊗1)†=σ:=∑jfj∣ajbj⟩⟨ajbj∣.\frac1D\sum_{k=0}^{D-1}(U_k\otimes\mathbf1)|\Phi\rangle\langle\Phi|(U_k\otimes\mathbf1)^\dagger=\sigma:=\sum_jf_j|a_jb_j\rangle\langle a_jb_j|.D1​k=0∑D−1​(Uk​⊗1)∣Φ⟩⟨Φ∣(Uk​⊗1)†=σ:=j∑​fj​∣aj​bj​⟩⟨aj​bj​∣.

This is a separable density matrix and its attained classical mutual information is

I(A:B)σ=H(f)=S(ΦA).I(A:B)_\sigma=H(f)=S(\Phi_A).I(A:B)σ​=H(f)=S(ΦA​).

The construction supplies an explicit finite disentangling map, including unequal ambient dimensions and D=1D=1D=1. It makes no optimality assertion over general local operations.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P47
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Intakenext100.P47

open QAlgorithms Filter

open QAlgorithms.Papers.Independent.Gap48.P20 (Copies localProduct)

open QAlgorithms.Papers.Independent.Dependency50.P19 (mutualInformation first)

open QAlgorithms.Papers.Independent.Dependency50.P45 (entropyExchange)

open QAlgorithms.Papers.Independent.Information100.P27 (shannon Config)

open QAlgorithms.Papers.Independent.Information100.P71 (channelPower)

open QAlgorithms.Papers.Independent.Advance50.P05 (RectangularInstrument)

open scoped Matrix ComplexConjugate

attribute [local instance] Classical.propDecidable Classical.decEq

/-- PDF p.8, III.C: D equiprobable phase unitaries, separability, and the attained classical bit entropy.
The index range 0,...,D-1 is the same root-of-unity ensemble as the printed 1,...,D. -/
theorem schmidtPhase_disentangling {A B : Type} [Fintype A] [Fintype B]
    [Nonempty A] [Nonempty B] (D : ℕ) (hD : 0 < D) (f : Fin D → ℝ)
    (hf : ∀ j, 0 < f j) (hsum : ∑ j, f j = 1)
    (a : Matrix A (Fin D) ℂ) (b : Matrix B (Fin D) ℂ)
    (ha : aᴴ * a = 1) (hb : bᴴ * b = 1) :
    ∃ R : COLUR A B, ∃ e : Fin R.count ≃ Fin D,
      R.count = D ∧ (∀ i, R.weight i = (D : ℝ)⁻¹) ∧ R.allowed .alice ∧
      (∀ i, (R.alice i : Matrix A A ℂ) = schmidtPhase a (e i)) ∧
      R.operation (pureDensity (schmidtKet f a b)) = schmidtDiagonal f a b ∧
      Entanglement.IsSeparableDensity (schmidtDiagonal f a b) ∧
      mutualInformation (schmidtDiagonal f a b) = shannon f ∧
      shannon f = vnEntropy (traceRight (pureDensity (schmidtKet f a b))) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P47

end
Source
On the quantum, classical and total amount of correlations in a quantum state, arXiv:quant-ph/0410091v2, PDF p. 8, Section III.C: Schmidt phase randomization and the attained classical correlation
Read-back

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

Let A,BA,BA,B be finite nonempty types, let D>0D>0D>0 be a natural number, let fj>0f_j>0fj​>0 for every j<Dj<Dj<D, with ∑jfj=1\sum_j f_j=1∑j​fj​=1, and let a∈CA×Da\in\mathbb C^{A\times D}a∈CA×D and b∈CB×Db\in\mathbb C^{B\times D}b∈CB×D satisfy a†a=b†b=IDa^\dagger a=b^\dagger b=I_Da†a=b†b=ID​. Write αj(x)=axj\alpha_j(x)=a_{xj}αj​(x)=axj​, βj(y)=byj\beta_j(y)=b_{yj}βj​(y)=byj​, and define

ψ(x,y)=∑j<Dfj axjbyj,Δ(x,y),(x′,y′)=∑j<Dfjaxjbyjax′jby′j‾.\psi(x,y)=\sum_{j<D}\sqrt{f_j}\,a_{xj}b_{yj}, \qquad \Delta_{(x,y),(x',y')} =\sum_{j<D}f_j a_{xj}b_{yj}\overline{a_{x'j}b_{y'j}}.ψ(x,y)=j<D∑​fj​​axj​byj​,Δ(x,y),(x′,y′)​=j<D∑​fj​axj​byj​ax′j​by′j​​.

Then there exist a finite random local-unitary family RRR and a bijection e:{0,…,R.count−1}→{0,…,D−1}e:\{0,\ldots,R.\mathrm{count}-1\}\to\{0,\ldots,D-1\}e:{0,…,R.count−1}→{0,…,D−1} such that R.count=DR.\mathrm{count}=DR.count=D, every weight is 1/D1/D1/D, every Bob unitary is the identity, and Alice’s unitary at index iii has entries

(Ui)xx′=δxx′+∑j<D(e2πi(e(i))j/D−1)axjax′j‾.(U_i)_{xx'}=\delta_{xx'}+ \sum_{j<D}\left(e^{2\pi i(e(i))j/D}-1\right)a_{xj}\overline{a_{x'j}}.(Ui​)xx′​=δxx′​+j<D∑​(e2πi(e(i))j/D−1)axj​ax′j​​.

Here a random local-unitary family consists of a positive natural count, nonnegative weights summing to one, and one unitary on each side at every index; its operation is R(X)=∑iwi(Ui⊗Vi)X(Ui⊗Vi)†\mathcal R(X)=\sum_i w_i(U_i\otimes V_i)X(U_i\otimes V_i)^\daggerR(X)=∑i​wi​(Ui​⊗Vi​)X(Ui​⊗Vi​)†. The asserted family satisfies R(∣ψ⟩⟨ψ∣)=Δ\mathcal R(|\psi\rangle\langle\psi|)=\DeltaR(∣ψ⟩⟨ψ∣)=Δ; moreover, Δ\DeltaΔ is a finite convex combination of tensor products of positive semidefinite trace-one matrices, and

S(Tr⁡BΔ)+S(Tr⁡AΔ)−S(Δ)=−∑j<Dfjlog⁡2fj=S(Tr⁡B∣ψ⟩⟨ψ∣).S(\operatorname{Tr}_B\Delta)+S(\operatorname{Tr}_A\Delta)-S(\Delta) =-\sum_{j<D}f_j\log_2 f_j =S(\operatorname{Tr}_B|\psi\rangle\langle\psi|).S(TrB​Δ)+S(TrA​Δ)−S(Δ)=−j<D∑​fj​log2​fj​=S(TrB​∣ψ⟩⟨ψ∣).

Here (∣ψ⟩⟨ψ∣)uv=ψ(u)ψ(v)‾(|\psi\rangle\langle\psi|)_{uv}=\psi(u)\overline{\psi(v)}(∣ψ⟩⟨ψ∣)uv​=ψ(u)ψ(v)​, and S(X)=−Re⁡tr⁡[(x↦xlog⁡2x)(X)]S(X)=-\operatorname{Re}\operatorname{tr}[(x\mapsto x\log_2x)(X)]S(X)=−Retr[(x↦xlog2​x)(X)], using continuous functional calculus and the convention 0log⁡20=00\log_2 0=00log2​0=0. The case D=1D=1D=1 is included; neither the family nor the bijection is asserted to be unique.

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