Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Bounded logarithmic rates and nonvanishing for the saddle-point forms

Open
ZudilinZeta.zudilin_saddle_log_side_conditions

by tomasz · Oct 1, 2026 · Mathlib 0df444a (Lean v4.33.1)

analysisnumber-theoryzeta-values

Let PPP be admissible with r=3r=3r=3, and let τ0\tau_0τ0​ satisfy the saddle-point hypotheses of Lemma 2: it is a zero of the characteristic polynomial, has positive imaginary part and maximal real part among such zeros, satisfies Re⁡τ0<η0\operatorname{Re}\tau_0<\eta_0Reτ0​<η0​, and obeys Im⁡f0(τ0)∉πZ\operatorname{Im}f_0(\tau_0)\notin\pi\mathbb ZImf0​(τ0​)∈/πZ.

For the linear forms FnF_nFn​ in (2), the logarithmic rates are eventually bounded above, and nonzero forms occur at arbitrarily large indices:

(∃B∈R  ∃N∈N  ∀n≥N:log⁡∣Fn∣n≤B)and(∀N∈N  ∃n≥N:Fn≠0).\left(\exists B\in\mathbb R\;\exists N\in\mathbb N\;\forall n\ge N:\quad \frac{\log|F_n|}{n}\le B\right) \quad\text{and}\quad \left(\forall N\in\mathbb N\;\exists n\ge N:\quad F_n\ne0\right).(∃B∈R∃N∈N∀n≥N:nlog∣Fn​∣​≤B)and(∀N∈N∃n≥N:Fn​=0).

These are analytic side conditions of the finite exponential-rate assertion obtained by the saddle-point method in Lemma 2. They permit the existing real-valued limsup statement to be used to produce arbitrarily small nonzero normalized forms.

Formalization Note Lean's real logarithm satisfies log⁡0=0\log 0=0log0=0, and its real-valued limsup alone does not certify boundedness or nonvanishing. The displayed logarithmic rate uses these totalized real operations, including division at n=0n=0n=0; the eventual statements ignore that initial index. Boundedness and nonvanishing are separate proof obligations here, not consequences asserted solely from the existing formal limsup equality.

Preamble
import Definitions.Def_ZudilinZetaAsymp
Formal statement
namespace ZudilinZeta

theorem zudilin_saddle_log_side_conditions (P : Params) (hr : P.r = 3) (τ₀ : ℂ)
    (hroot : charPoly P τ₀ = 0) (him : 0 < τ₀.im)
    (hmax : ∀ τ : ℂ, charPoly P τ = 0 → 0 < τ.im → τ.re ≤ τ₀.re)
    (hre : τ₀.re < (P.eta 0 : ℝ))
    (hpi : ∀ k : ℤ, (f0 P τ₀).im ≠ (k : ℝ) * Real.pi) :
    Filter.IsBoundedUnder (fun x y : ℝ => x ≤ y) Filter.atTop
        (fun n : ℕ => Real.log |F P n| / (n : ℝ)) ∧
      (∃ᶠ n : ℕ in Filter.atTop, F P n ≠ 0) := by sorry

end ZudilinZeta
Source
W. V. Zudilin, One of the numbers ζ(5), ζ(7), ζ(9), ζ(11) is irrational, Russian Math. Surveys 56:4 (2001), 774–776, p. 775; https://doi.org/10.1070/RM2001v056n04ABEH000427; full text https://www.mathnet.ru/php/getFT.phtml?jrnid=rm&option_lang=eng&paperid=427&what=fullteng. Lemma 2 and the immediately following passage from (3) to (4); explicit side conditions of the saddle-point asymptotic.

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