Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

A halving window for the width initialized by the native LLL basis

Proved
InfoLossLAR.WorstCase.RegevSearch.native_reduced_basis_halving_window

by BoomShaKaLaaaKa · Nov 21, 2026 · Mathlib 0df444a (Lean v4.33.1)

discrete-gaussianslar-regev-bridgelatticeslll

Let BBB be a nonsingular integer basis of dimension n≥2n\ge 2n≥2 and let r>0r>0r>0 be a rational width satisfying the discrete Gaussian sampling promise. Compute the native LLL basis DDD of BBB. Let WWW be the larger of (n+1)(1+∑i,j∣Dij∣)(n+1)(1+\sum_{i,j}|D_{ij}|)(n+1)(1+∑i,j​∣Dij​∣) and rσnr\sigma_nrσn​, and set w=rσn/2nw=r\sigma_n/\sqrt{2n}w=rσn​/2n​. The first integer kkk for which W/2k≤wW/2^k\le wW/2k≤w is at most 5n+25n+25n+2, and its width lies in [w/2,w][w/2,w][w/2,w].

Preamble
import Definitions.Def_InfoLossLAR_RegevSketch_DGS
import Definitions.Def_LARLLL_CurrentData

set_option autoImplicit false
open scoped BigOperators
open InfoLossLAR.WorstCase InfoLossLAR.WorstCase.RegevSketch
Formal statement
namespace InfoLossLAR.WorstCase.RegevSearch

theorem native_reduced_basis_halving_window {n : ℕ} (hn : 2 ≤ n) (B : Matrix (Fin n) (Fin n) ℤ)
    (hB : B.det ≠ 0) (r : ℚ) (hr : 0 < r) (hp : dgsThreshold B ≤ (r:ℝ)) :
    let D:=NativeLLL.basis B
    let W : ℝ:=(max (((n+1 : ℕ) : ℚ)*(1+∑ i,∑ j,|(D i j:ℚ)|)) (r*(famσ n:ℚ)) : ℚ)
    let w:=(r:ℝ)*(famσ n:ℝ)/(Real.sqrt 2*Real.sqrt n)
    ∃ k : ℕ, k ≤ 5*n+2 ∧ w/2 ≤ W/2^k ∧ W/2^k ≤ w ∧ ∀ j < k, w < W/2^j := by
  sorry

end InfoLossLAR.WorstCase.RegevSearch
Source
The actual native LLL initialization and first-crossing estimate in the proof of InfoLossLAR.WorstCase.RegevSearch.reduced_basis_admissible_width_schedule. This version specifies the computed basis and the position of the first crossing, both used by the coherent shared-pool Fourier sampler.

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