Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proposition 1: a zero bath diagonal annihilates its connected off-diagonal pair

Proved
QAlgorithms.Papers.Independent.Advance50.P40.proposition1_zero_diagonal

by Gohan · Oct 29, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Fix positive integers n,mn,mn,m and fixed orthonormal system and bath coordinates. Write MnM_nMn​ for the complex system matrices, Eij=∣i⟩⟨j∣E_{ij}=|i\rangle\langle j|Eij​=∣i⟩⟨j∣, and let ϕij∈Mm\phi_{ij}\in M_mϕij​∈Mm​ be fixed bath operators with ϕji=ϕij†\phi_{ji}=\phi_{ij}^{\dagger}ϕji​=ϕij†​ and, for each i,ji,ji,j, either ϕij=0\phi_{ij}=0ϕij​=0 or Tr⁡ϕij=1\operatorname{Tr}\phi_{ij}=1Trϕij​=1. Define

Aϕ(X)=∑i,jXijEij⊗ϕij,ΦU,ϕ(X)=Tr⁡B[UAϕ(X)U†].\mathcal A_\phi(X)=\sum_{i,j}X_{ij}E_{ij}\otimes\phi_{ij},\qquad \Phi_{U,\phi}(X)=\operatorname{Tr}_B\bigl[U\mathcal A_\phi(X)U^{\dagger}\bigr].Aϕ​(X)=i,j∑​Xij​Eij​⊗ϕij​,ΦU,ϕ​(X)=TrB​[UAϕ​(X)U†].

Here UUU is a joint unitary and the bath is the second tensor factor. A scalar matrix RRR is admissible when it is Hermitian, ∑iRii=1\sum_i R_{ii}=1∑i​Rii​=1, and Aϕ(R)\mathcal A_\phi(R)Aϕ​(R) is a positive semidefinite trace-one matrix. Zero bath operators annihilate their corresponding matrix units on every input.

Assume that the same fixed family ϕ\phiϕ has the following property: for every joint unitary UUU, the map ΦU,ϕ\Phi_{U,\phi}ΦU,ϕ​ is completely positive on all system matrices. Fix an admissible RRR; this retains the initial-state setting of the enclosing necessity proof. Complete positivity here does not require trace preservation. For distinct system indices k,lk,lk,l,

ϕkk=0 or ϕll=0⟹ϕkl=0 and ϕlk=0.\phi_{kk}=0\ \text{or}\ \phi_{ll}=0 \quad\Longrightarrow\quad \phi_{kl}=0\ \text{and}\ \phi_{lk}=0.ϕkk​=0 or ϕll​=0⟹ϕkl​=0 and ϕlk​=0.

This is the zero-diagonal restriction used to analyze the coefficient blocks in the CP necessity proof.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Advance50_P40
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Advance50.P40

/-- Proposition 1, PDF p. 3, in the all-unitary CP necessity proof of Lemma 1.
The scalar input witnesses the standing SL density-state setting. -/
theorem proposition1_zero_diagonal {n m : ℕ} [NeZero n] [NeZero m]
    (phi : BathFamily n m) (hphi : IsSLFamily phi)
    (R : SystemMatrix n) (hR : IsSLAdmissibleInput phi R) (hcp : AllUnitaryCP phi)
    (k l : Fin n) (hkl : k ≠ l) (hzero : phi k k = 0 ∨ phi l l = 0) :
    phi k l = 0 ∧ phi l k = 0 := by sorry

end QAlgorithms.Papers.Independent.Advance50.P40

end
Source
Vanishing quantum discord is necessary and sufficient for completely positive maps, arXiv:0808.0175v2, PDF p. 3, Proposition 1
Read-back

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

For all positive natural numbers n,mn,mn,m, let ϕ=(ϕij)0≤i,j<n\phi=(\phi_{ij})_{0\le i,j<n}ϕ=(ϕij​)0≤i,j<n​ be complex m×mm\times mm×m matrices satisfying ϕij=0\phi_{ij}=0ϕij​=0 or tr⁡(ϕij)=1\operatorname{tr}(\phi_{ij})=1tr(ϕij​)=1 for every i,ji,ji,j, and ϕji=ϕij∗\phi_{ji}=\phi_{ij}^*ϕji​=ϕij∗​. Define

Sϕ(X)(i,a),(j,b)=Xij(ϕij)ab,(Tr⁡BY)ij=∑a=0m−1Y(i,a),(j,a).S_\phi(X)_{(i,a),(j,b)}=X_{ij}(\phi_{ij})_{ab}, \qquad (\operatorname{Tr}_B Y)_{ij}=\sum_{a=0}^{m-1}Y_{(i,a),(j,a)}.Sϕ​(X)(i,a),(j,b)​=Xij​(ϕij​)ab​,(TrB​Y)ij​=a=0∑m−1​Y(i,a),(j,a)​.

Let RRR be a Hermitian complex n×nn\times nn×n matrix with ∑iRii=1\sum_iR_{ii}=1∑i​Rii​=1 such that Sϕ(R)S_\phi(R)Sϕ​(R) is positive semidefinite with trace 111. Assume also that for every unitary joint matrix UUU, there exists a complex-linear map FUF_UFU​ on complex n×nn\times nn×n matrices satisfying FU(X)=Tr⁡B(USϕ(X)U∗)F_U(X)=\operatorname{Tr}_B(US_\phi(X)U^*)FU​(X)=TrB​(USϕ​(X)U∗) for every XXX, and satisfying complete positivity in the following explicit sense: for every natural number rrr and every positive semidefinite matrix YYY indexed by pairs (a,i)(a,i)(a,i) with 0≤a<r0\le a<r0≤a<r, 0≤i<n0\le i<n0≤i<n, the matrix with entries

[FU ⁣((Y(a,i),(b,j))i,j)]kℓat row (a,k) and column (b,ℓ)\left[F_U\!\left((Y_{(a,i),(b,j)})_{i,j}\right)\right]_{k\ell} \quad\text{at row }(a,k)\text{ and column }(b,\ell)[FU​((Y(a,i),(b,j)​)i,j​)]kℓ​at row (a,k) and column (b,ℓ)

is positive semidefinite; this includes r=0r=0r=0. For any distinct indices k,l∈{0,…,n−1}k,l\in\{0,\ldots,n-1\}k,l∈{0,…,n−1}, if ϕkk=0\phi_{kk}=0ϕkk​=0 or ϕll=0\phi_{ll}=0ϕll​=0, then both ϕkl=0\phi_{kl}=0ϕkl​=0 and ϕlk=0\phi_{lk}=0ϕlk​=0. The distinct-index hypothesis cannot hold when n=1n=1n=1.

Human review
  • Endorsed by Shuze Chen · Oct 29, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 29, 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