Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lean source theorem: sq_tsum_le

Proved
ErdosProblems.Shared.DirichletPole.sq_tsum_le

by willcook · Sep 25, 2026 · Mathlib c5ea003 (Lean v4.30.0)

lean-sharedsupport-lemma

Under the stated nonnegative Dirichlet-convolution bounds and a uniform bound B on weighted partial sums of b, the square of the Dirichlet series sum for a at any real s≥1 is at most B² times the corresponding sum for m, provided both series are summable.

Preamble
import Definitions.Def_ErdosProblems_Shared_DirichletPoleComparison
import Mathlib.Analysis.SpecialFunctions.Pow.Real
import Mathlib.NumberTheory.LSeries.Convolution

/-!
# Two simple poles exclude a squared comparison of Dirichlet series

Let `a, g, b, m : ℕ → ℝ` be nonnegative with

* `a ≤ g ⋆ b` and `g ⋆ g ≤ m` coefficientwise (Dirichlet convolution, `n ≠ 0`),
* `g ≤ a`, and the partial sums of `∑ b(n) / n` bounded by `B`.

If the Dirichlet series of `a` and of `m` both have a simple pole at `s = 1`
(`(s - 1) L(a, s) → r_a > 0` and `(s - 1) L(m, s) → r_m ≠ 0` as `s → 1⁺`), this is
impossible.  Indeed, for real `s > 1` near `1`,

`L(a, s) ^ 2 ≤ (L(g, s) L(b, s)) ^ 2 ≤ B ^ 2 L(g ⋆ g, s) ≤ B ^ 2 L(m, s)`,

so `((s - 1) L(a, s)) ^ 2 ≤ B ^ 2 (s - 1) · (s - 1) L(m, s)`; the left side tends to
`r_a ^ 2 > 0` and the right side to `0`.

This is the analytic core of the classical proof that a non-square element of a number
field is a non-square modulo infinitely many degree-one primes: `a` counts the ideals of
the base field, `m` those of the quadratic extension, `g` the ideals built from split
degree-one primes and `b` the rest.  Summability of the two series near `s = 1` is not
assumed: it follows from the nonzero limits, since a divergent series has `tsum = 0`.
-/

noncomputable section


open Filter Topology
open scoped LSeries.notation

open ErdosProblems.Shared.DirichletPole

Formal statement
theorem ErdosProblems.Shared.DirichletPole.sq_tsum_le {a g b m : ℕ → ℝ}
    (ha : ∀ n, 0 ≤ a n) (hg : ∀ n, 0 ≤ g n) (hb : ∀ n, 0 ≤ b n)
    (hga : ∀ n, g n ≤ a n)
    (h1 : ∀ n, n ≠ 0 → a n ≤ dconv g b n)
    (h2 : ∀ n, n ≠ 0 → dconv g g n ≤ m n)
    (B : ℝ) (h3 : ∀ X : ℕ, ∑ n ∈ Finset.range X, b n / n ≤ B)
    {s : ℝ} (hs : 1 ≤ s) (hsa : Summable (rterm a s)) (hsm : Summable (rterm m s)) :
    (∑' n, rterm a s n) ^ 2 ≤ B ^ 2 * ∑' n, rterm m s n := by sorry
Source
Lean source (Apache-2.0): https://github.com/wcook04/plectis-erdos-lean/blob/4fe59e0191169606942baae0949365070b7f419c/ErdosProblems/Shared/DirichletPoleComparison.lean#L108-L143 Related paper by Will Cook (CC-BY-4.0): https://github.com/wcook04/plectis-erdos/blob/6917e15ec4abc2623512254da93221e446eeb707/paper/243/erdos-243-reciprocal-tail-rigidity.tex#L1-L58 Paper's AI-assistance disclosure: https://github.com/wcook04/plectis-erdos/blob/6917e15ec4abc2623512254da93221e446eeb707/paper/paper-house-style.sty#L180-L188 Original Erdős problem and Koizumi prior work are distinguished in the paper: https://github.com/wcook04/plectis-erdos/blob/6917e15ec4abc2623512254da93221e446eeb707/paper/243/erdos-243-reciprocal-tail-rigidity.tex#L57-L110

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