Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Unconditional geometric overlap decoding bound

Open
QAlgorithms.Papers.Independent.Intakenext100.P08.geometricOverlap_error

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

prove2quantum

Fix N≥1N\ge1N≥1, coordinate noise variance v>0v>0v>0, radius a>0a>0a>0, and cell volume Δ>0\Delta>0Δ>0. There is a real symplectic matrix SSS such that the row lattice Λ=L(Δ1/(2N)S)\Lambda=L(\Delta^{1/(2N)}S)Λ=L(Δ1/(2N)S) has covolume Δ\DeltaΔ and, for a random vector ξ∈R2N\xi\in\mathbb R^{2N}ξ∈R2N with independent N(0,v)N(0,v)N(0,v) coordinates,

Pr⁡{ξ∉Vor⁡0(Λ)}≤Pr⁡{∥ξ∥>a}+V2Na2NΔ.\Pr\{\xi\notin\operatorname{Vor}_0(\Lambda)\}\le\Pr\{\|\xi\|>a\}+\frac{V_{2N}a^{2N}}{\Delta}.Pr{ξ∈/Vor0​(Λ)}≤Pr{∥ξ∥>a}+ΔV2N​a2N​.

The lattice is chosen before the displacement is sampled. The central Voronoi cell uses the Euclidean norm; boundaries have zero Gaussian measure. The first term restores the atypical shifts expressly omitted in (59)–(64). This bounds a geometric decoding event and does not establish quantum-code fidelity.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P08
Formal statement
namespace QAlgorithms.Papers.Independent.Intakenext100.P08

open MeasureTheory ProbabilityTheory Matrix

open scoped BigOperators ENNReal

-- Accepted grouped index and source-sign matrix; no paper-local copies.

open QAlgorithms.Papers.Independent.Dependency50.P10 (Phase sigma)
-- The accepted predicate uses column bases and -sigma: apply it to S.transpose.

open QAlgorithms.Papers.Independent.Frontier30.P02 (IsSymplectic)
-- The accepted pairing has the opposite sign: exchange its two arguments below.

open QAlgorithms.Papers.Independent.Frontier30.P10 (symplecticPair)

/-- PDF pp.6–7, (58)–(64): unconditional geometric overlap bound.
Printed/changed/why: (61) explicitly ignores shifts larger than a; add their
Gaussian tail to bound the full failure probability. This is not physical-code achievability. -/
theorem geometricOverlap_error (N : ℕ) (hN : 1 ≤ N) (v : NNReal) (hv : 0 < v)
    (a Delta : ℝ) (ha : 0 < a) (hDelta : 0 < Delta) :
    ∃ S : Generator N, IsSymplectic S.transpose ∧ S.det ≠ 0 ∧
      let M : Generator N := Delta ^ (1 / (2 * (N : ℝ))) • S
      |M.det| = Delta ∧
      decodingFailure M v ≤ displacementTail N v a +
        unitBallVolume (2 * N) * a ^ (2 * N) / Delta := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P08
Source
Achievable rates for the Gaussian quantum channel, arXiv:quant-ph/0105058v1, PDF p. 6, §V, (58)–(64): overlap decoding bound with the omitted tail restored
Read-back

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

For every natural number N≥1N\ge1N≥1, every strictly positive nonnegative real number vvv, and all positive real numbers aaa and Δ\DeltaΔ, there exists a real 2N×2N2N\times2N2N×2N matrix SSS such that SΩNST=ΩNS\Omega_NS^{\mathsf T}=\Omega_NSΩN​ST=ΩN​ and det⁡S≠0\det S\ne0detS=0, where ΩN=(0IN−IN0)\Omega_N=\begin{pmatrix}0&I_N\\-I_N&0\end{pmatrix}ΩN​=(0−IN​​IN​0​). Set M=Δ1/(2N)SM=\Delta^{1/(2N)}SM=Δ1/(2N)S, using the real power, let LML_MLM​ be the set of all integer linear combinations of its rows, and define

VM={x∈R2N:∥x∥≤∥x−y∥ for every y∈LM}.V_M=\{x\in\mathbb R^{2N}:\|x\|\le\|x-y\|\text{ for every }y\in L_M\}.VM​={x∈R2N:∥x∥≤∥x−y∥ for every y∈LM​}.

Let γN,v\gamma_{N,v}γN,v​ be the probability measure on Euclidean R2N\mathbb R^{2N}R2N obtained from 2N2N2N independent real Gaussian coordinates, each with mean 000 and variance vvv, and let c2Nc_{2N}c2N​ be the real Lebesgue volume of the open unit ball in R2N\mathbb R^{2N}R2N. Then

∣det⁡M∣=Δ,γN,v(R2N∖VM)≤γN,v{x:a<∥x∥}+c2Na2NΔ.|\det M|=\Delta,\qquad \gamma_{N,v}(\mathbb R^{2N}\setminus V_M) \le \gamma_{N,v}\{x:a<\|x\|\} +\frac{c_{2N}a^{2N}}{\Delta}.∣detM∣=Δ,γN,v​(R2N∖VM​)≤γN,v​{x:a<∥x∥}+Δc2N​a2N​.

The defining inequalities for VMV_MVM​ include equality, while the radial tail uses the strict inequality a<∥x∥a<\|x\|a<∥x∥.

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