Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Independent edge Bell products have additive trace-norm error

Open
QAlgorithms.Papers.Independent.Intakenext100.P84.independent_edge_bell_error

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

prove2quantum

Let EEE be the finite set of labelled edges. For every e∈Ee\in Ee∈E, let me∈Nm_e\in\mathbb Nme​∈N and let ρe\rho_eρe​ be a normalized density operator on the two endpoint registers of mem_eme​ Bell pairs. Write Φme=(∣Φ+⟩⟨Φ+∣)⊗me\Phi_{m_e}=(|\Phi^+\rangle\langle\Phi^+|)^{\otimes m_e}Φme​​=(∣Φ+⟩⟨Φ+∣)⊗me​, where ∣Φ+⟩=(∣00⟩+∣11⟩)/2|\Phi^+\rangle=(|00\rangle+|11\rangle)/\sqrt2∣Φ+⟩=(∣00⟩+∣11⟩)/2​. Suppose ε>0\varepsilon>0ε>0 and ∥ρe−Φme∥1≤ε\|\rho_e-\Phi_{m_e}\|_1\le\varepsilon∥ρe​−Φme​​∥1​≤ε for every edge. Then

∥⨂e∈Eρe−⨂e∈EΦme∥1≤∣E∣ε.\left\|\bigotimes_{e\in E}\rho_e-\bigotimes_{e\in E}\Phi_{m_e}\right\|_1\le |E|\varepsilon.​e∈E⨂​ρe​−e∈E⨂​Φme​​​1​≤∣E∣ε.

The norm is the full Schatten 111-norm. Empty products are the one-dimensional unit state, and zero Bell copies are allowed. This estimate controls the error of the independently prepared Bell-resource network.

Formalization Note The two halves of each Bell product are grouped by endpoint, so its vector is the accepted normalized maximally entangled vector on two mem_eme​-qubit registers.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P84
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Intakenext100.P84

open QAlgorithms

open scoped BigOperators

attribute [local instance] Classical.propDecidable Classical.decEq

/-- PDF p.3, Eq.(4): independent normalized edge outputs, each epsilon-close
in full trace norm to m(e) Bell pairs, have joint error at most |E| epsilon.
`maxEntangled m` is the Bell product with the two endpoint registers grouped
locally; its amplitude is 2^(-m/2). Zero copies give the tensor unit. -/
theorem independent_edge_bell_error {E : Type*} [Fintype E]
    (m : E → ℕ)
    (rho : (e : E) → Matrix (Qubits (m e) × Qubits (m e))
      (Qubits (m e) × Qubits (m e)) ℂ)
    (epsilon : ℝ) (hepsilon : 0 < epsilon)
    (hrho : ∀ e, IsDensityMatrix (rho e))
    (hedge : ∀ e,
      traceNorm (rho e - pureDensity (maxEntangled (m e))) ≤ epsilon) :
    traceNorm (edgeTensor rho -
      edgeTensor (fun e => pureDensity (maxEntangled (m e)))) ≤
        (Fintype.card E : ℝ) * epsilon := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P84

end
Source
Aggregating quantum repeaters for the quantum internet, arXiv:1606.00135v1, PDF p. 3, Eq.(4)
Read-back

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

For every finite type EEE, every function m:E→Nm:E\to\mathbb Nm:E→N, every family of complex matrices ρe\rho_eρe​ indexed by pairs of mem_eme​-bit strings, and every real ϵ>0\epsilon>0ϵ>0, suppose that each ρe\rho_eρe​ is positive semidefinite with trace 111, and that ∥ρe−Be∥tr≤ϵ\|\rho_e-B_e\|_{\mathrm{tr}}\le\epsilon∥ρe​−Be​∥tr​≤ϵ for every e∈Ee\in Ee∈E. Here an rrr-bit string is a function {0,…,r−1}→{false,true}\{0,\ldots,r-1\}\to\{\mathrm{false},\mathrm{true}\}{0,…,r−1}→{false,true}, BeB_eBe​ has entries Be(u,v)=Φe(u)Φe(v)‾B_e(u,v)=\Phi_e(u)\overline{\Phi_e(v)}Be​(u,v)=Φe​(u)Φe​(v)​, and, for a pair u=(a,b)u=(a,b)u=(a,b) of mem_eme​-bit strings, Φe(a,b)=1/2me\Phi_e(a,b)=1/\sqrt{2^{m_e}}Φe​(a,b)=1/2me​​ when a=ba=ba=b and 000 otherwise. For any complex square matrix AAA, the notation ∥A∥tr\|A\|_{\mathrm{tr}}∥A∥tr​ means Re⁡tr⁡A†A\operatorname{Re}\operatorname{tr}\sqrt{A^\dagger A}RetrA†A​, with the square root taken by continuous functional calculus. Define matrices Tρ,TBT_\rho,T_BTρ​,TB​, indexed by tuples of such pairs, by (Tρ)x,y=∏e∈Eρe(xe,ye)(T_\rho)_{x,y}=\prod_{e\in E}\rho_e(x_e,y_e)(Tρ​)x,y​=∏e∈E​ρe​(xe​,ye​) and (TB)x,y=∏e∈EBe(xe,ye)(T_B)_{x,y}=\prod_{e\in E}B_e(x_e,y_e)(TB​)x,y​=∏e∈E​Be​(xe​,ye​). Then

∥Tρ−TB∥tr≤∣E∣ϵ.\|T_\rho-T_B\|_{\mathrm{tr}}\le |E|\epsilon.∥Tρ​−TB​∥tr​≤∣E∣ϵ.

The values me=0m_e=0me​=0 are included, in which case the corresponding pair index set is a singleton and Be=[1]B_e=[1]Be​=[1]. If EEE is empty, both product matrices are [1][1][1], and the conclusion is 0≤00\le00≤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