Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Second-moment bound conditional on geometric sums

Proved
LL_Decomp1543a.square_expand

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

number-theory

For real c,αc,\alphac,α, natural NNN, U≥0U\ge0U≥0 and M≥1M\ge1M≥1, assume ∣∑A<t≤Be(θt)∣≤K(M+1/2,θ)\left|\sum_{A<t\le B}e(\theta t)\right|\le K(M+1/2,\theta)​∑A<t≤B​e(θt)​≤K(M+1/2,θ) for every real θ\thetaθ and integers A,BA,BA,B with ⌊M⌋≤A\lfloor M\rfloor\le A⌊M⌋≤A and B≤⌊2M⌋B\le\lfloor2M\rfloorB≤⌊2M⌋. Conditional on that hypothesis, ∑t∈T(N,M)∣I(c,N,U,α,t)∣2≤∑m∈L(N,U,N/M)am(U)2∑l∈L(N,U,N/M)K(M+1/2,α(m−l))\sum_{t\in T(N,M)}|I(c,N,U,\alpha,t)|^2\le\sum_{m\in L(N,U,N/M)}a_m(U)^2\sum_{l\in L(N,U,N/M)}K(M+1/2,\alpha(m-l))∑t∈T(N,M)​∣I(c,N,U,α,t)∣2≤∑m∈L(N,U,N/M)​am​(U)2∑l∈L(N,U,N/M)​K(M+1/2,α(m−l)).

Preamble
import Mathlib
import Definitions.Def_LL_Decomp1543a_e
import Definitions.Def_LL_Decomp1543a_nint
import Definitions.Def_LL_Decomp1543a_minK
import Definitions.Def_LL_Decomp1543a_am
import Definitions.Def_LL_Decomp1543a_tBlock
import Definitions.Def_LL_Decomp1543a_mRange
import Definitions.Def_LL_Decomp1543a_inner
import Definitions.Def_LL_Decomp1543a_lRange

open Real Finset ArithmeticFunction
Formal statement
namespace LL_Decomp1543a
theorem square_expand (c : ℝ) (N : ℕ) (U M α : ℝ) (hU : 0 ≤ U) (hM : 1 ≤ M)
    (hgeom : ∀ θ : ℝ, ∀ A B : ℤ, ⌊M⌋ ≤ A → B ≤ ⌊2 * M⌋ →
      ‖∑ t ∈ Ioc A B, e (θ * t)‖ ≤ minK (M + 1 / 2) θ) :
    ∑ t ∈ tBlock N M, ‖inner c N U α t‖ ^ 2 ≤
      ∑ m ∈ lRange N U (N / M), am U m ^ 2 *
        ∑ l ∈ lRange N U (N / M), minK (M + 1 / 2) (α * ((m : ℝ) - l)) := by sorry
end LL_Decomp1543a
Source
M. C. Liu and T. Z. Wang, On the Vinogradov bound in the three primes Goldbach conjecture, Acta Arith. 105, 2002, p. 171, first display following equation 6.11. https://doi.org/10.4064/aa105-2-3
Read-back

What the Lean code literally says, in plain math · claude-haiku

For real c,U,M,αc, U, M, \alphac,U,M,α, a natural number NNN, and hypotheses 0≤U0 \le U0≤U and 1≤M1 \le M1≤M, the statement assumes hgeom: for every real θ\thetaθ and all integers A,BA, BA,B with ⌊M⌋≤A\lfloor M \rfloor \le A⌊M⌋≤A and B≤⌊2M⌋B \le \lfloor 2M \rfloorB≤⌊2M⌋, the norm of ∑t∈(A,B]∩Ze(θt)\sum_{t \in (A,B]\cap\mathbb{Z}} e(\theta t)∑t∈(A,B]∩Z​e(θt) is at most minK(M+12,θ)\mathrm{minK}(M+\tfrac12,\theta)minK(M+21​,θ). When B≤AB \le AB≤A the sum is empty and equals 000. The conclusion is

∑t∈tBlock(N,M)∣inner(c,N,U,α,t)∣2≤∑m∈Lam(U,m)2∑l∈LminK(M+12, α(m−l)),\sum_{t \in \mathrm{tBlock}(N,M)} \bigl|\mathrm{inner}(c,N,U,\alpha,t)\bigr|^2 \le \sum_{m \in L} \mathrm{am}(U,m)^2 \sum_{l \in L} \mathrm{minK}\bigl(M+\tfrac12,\ \alpha(m-l)\bigr),t∈tBlock(N,M)∑​​inner(c,N,U,α,t)​2≤m∈L∑​am(U,m)2l∈L∑​minK(M+21​, α(m−l)),

where L=lRange(N,U,N/M)L = \mathrm{lRange}(N,U,N/M)L=lRange(N,U,N/M), N/MN/MN/M is real division, and m,lm, lm,l are indices cast to reals. The statement puts no constraint on ccc or α\alphaα beyond what hgeom and the conclusion involve. Since M≥1M \ge 1M≥1, no division by zero occurs. The proof is a placeholder (sorry), so the item asserts this without proof.

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

    Confirmed by the moderator at approval.

  • Endorsed by sthubbar · 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