Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Lemma 2.2 — Q(q1,q2,s,t)/Q(q1,q2,s−k1,t+k2)∈[1,(1+μ)3]Q(q_1,q_2,s,t)/Q(q_1,q_2,s-k_1,t+k_2)\in[1,(1+\mu)^3]Q(q1​,q2​,s,t)/Q(q1​,q2​,s−k1​,t+k2​)∈[1,(1+μ)3] (typo corrected)

Proved
ExplicitExpanders.Sizes.Q_ratio_step

by mikedeng1 · Oct 5, 2026 · Mathlib 0df444a (Lean v4.33.1)

expander-graphsnumber-theoryp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

Let q1q_1q1​ and q2q_2q2​ be distinct primes, let μ>0\mu>0μ>0, and let k1≥1k_1\ge1k1​≥1 and k2≥0k_2\ge0k2​≥0 be integers with

1≤q1k1q2k2≤1+μ.1\le\frac{q_1^{k_1}}{q_2^{k_2}}\le 1+\mu .1≤q2k2​​q1k1​​​≤1+μ.

Then for all integers s>k1s>k_1s>k1​ and t≥1t\ge1t≥1,

1≤Q(q1,q2,s,t)Q(q1,q2,s−k1,t+k2)≤(1+μ)3,1\le\frac{Q(q_1,q_2,s,t)}{Q(q_1,q_2,s-k_1,t+k_2)}\le(1+\mu)^3 ,1≤Q(q1​,q2​,s−k1​,t+k2​)Q(q1​,q2​,s,t)​≤(1+μ)3,

where QQQ is the LPS vertex count Q(q1,q2,s,t)=q13(s−1)q23(t−1)q1(q1−1)(q1+1)2q2(q2−1)(q2+1)2Q(q_1,q_2,s,t)=q_1^{3(s-1)}q_2^{3(t-1)}\frac{q_1(q_1-1)(q_1+1)}{2}\frac{q_2(q_2-1)(q_2+1)}{2}Q(q1​,q2​,s,t)=q13(s−1)​q23(t−1)​2q1​(q1​−1)(q1​+1)​2q2​(q2​−1)(q2​+1)​.

This is the last sentence of the proof of Lemma 2.2, with a misprint corrected. Moving from (s,t)(s,t)(s,t) to (s−k1,t+k2)(s-k_1,t+k_2)(s−k1​,t+k2​) divides QQQ by a factor in [1,(1+μ)3][1,(1+\mu)^3][1,(1+μ)3], so a chain of such moves passes through every interval [n,(1+μ)3n][n,(1+\mu)^3n][n,(1+μ)3n] with nnn large.

Formalization Note The paper prints Q(q1,q2,s−k1,t−k2)Q(q_1,q_2,s-k_1,t-k_2)Q(q1​,q2​,s−k1​,t−k2​) and "s,t≥max⁡{k1,k2}s,t\ge\max\{k_1,k_2\}s,t≥max{k1​,k2​}". As printed the sentence is false: Q(q1,q2,s,t)/Q(q1,q2,s−k1,t−k2)=q13k1q23k2Q(q_1,q_2,s,t)/Q(q_1,q_2,s-k_1,t-k_2)=q_1^{3k_1}q_2^{3k_2}Q(q1​,q2​,s,t)/Q(q1​,q2​,s−k1​,t−k2​)=q13k1​​q23k2​​, which is large. The intended pair is (s−k1,t+k2)(s-k_1,t+k_2)(s−k1​,t+k2​), for which the ratio is (q1k1/q2k2)3(q_1^{k_1}/q_2^{k_2})^3(q1k1​​/q2k2​​)3. The index conditions are adjusted: s>k1s>k_1s>k1​ makes s−k1s-k_1s−k1​ positive, and t≥1t\ge1t≥1; no lower bound on ttt in terms of k2k_2k2​ is needed after correcting the sign. Distinctness, μ>0\mu>0μ>0, and k1>0k_1>0k1​>0 are retained from the proof even though the algebraic identity alone needs none of them. The ratio is computed in R\mathbb RR after casting the natural numbers Q(⋅)Q(\cdot)Q(⋅).

Preamble
import Mathlib
import Definitions.Def_ExplicitExpanders_Sizes_Q
Formal statement
namespace ExplicitExpanders.Sizes

/-- Proof of Lemma 2.2 (arXiv:2003.11673v1, p. 7), last sentence, with the printed
`Q(q₁, q₂, s − k₁, t − k₂)` corrected to `Q(q₁, q₂, s − k₁, t + k₂)`: if
for distinct primes, `μ > 0`, `k₁ > 0`, and `1 ≤ q₁^{k₁} / q₂^{k₂} ≤ 1 + μ`, then for all
`s > k₁` and `t ≥ 1` the ratio between
`Q(q₁, q₂, s, t)` and `Q(q₁, q₂, s − k₁, t + k₂)` lies between `1` and `(1 + μ)³`. -/
theorem Q_ratio_step {q₁ q₂ : ℕ} (hq₁ : q₁.Prime) (hq₂ : q₂.Prime) (hne : q₁ ≠ q₂)
    {μ : ℝ} (hμ : 0 < μ) {k₁ k₂ : ℕ} (hk₁ : 0 < k₁)
    (hlow : 1 ≤ (q₁ : ℝ) ^ k₁ / (q₂ : ℝ) ^ k₂) (hupp : (q₁ : ℝ) ^ k₁ / (q₂ : ℝ) ^ k₂ ≤ 1 + μ)
    {s t : ℕ} (hs : k₁ < s) (ht : 1 ≤ t) :
    1 ≤ (Q q₁ q₂ s t : ℝ) / (Q q₁ q₂ (s - k₁) (t + k₂) : ℝ) ∧
      (Q q₁ q₂ s t : ℝ) / (Q q₁ q₂ (s - k₁) (t + k₂) : ℝ) ≤ (1 + μ) ^ 3 := by sorry

end ExplicitExpanders.Sizes
Source
N. Alon, Explicit expanders of every degree and size, arXiv:2003.11673v1, p. 7, proof of Lemma 2.2 (last sentence; printed t − k2 corrected to t + k2)
Human review
  • Endorsed by Shuze Chen · Oct 5, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 5, 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