Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Roy's Th. 4 over ℚ, semistable case, from the multiplicity estimate: if d dim(U ∩ S) ≤ n dim S for all S rational over ℚ and n = dim U < d, then (d − n)ℓ ≤ dn for ℚ-free z in U ∩ (ℚ + ℒ)ᵈ

Proved
DiazModulus.roy_rat_semistable_node

by carlok · Nov 14, 2026 · Mathlib 0df444a (Lean v4.33.1)

diaz-modulus-leannumber-theorytranscendence

Here L\mathcal LL is the set of logarithms of algebraic numbers (z∈Cz \in \mathbb Cz∈C with eze^zez algebraic). Write Q+L\mathbb Q + \mathcal LQ+L for the Q\mathbb QQ-vector space spanned by 111 and L\mathcal LL; it consists of the numbers q+λq + \lambdaq+λ with q∈Qq \in \mathbb Qq∈Q and λ∈L\lambda \in \mathcal Lλ∈L. A vector of Cd\mathbb C^dCd is rational if its coordinates are rational numbers, and a subspace S⊆CdS \subseteq \mathbb C^dS⊆Cd is rational over Q\mathbb QQ if it is spanned by its rational vectors.

Assume the multiplicity estimate hME, and assume the conclusion of DiazModulus.roy_rat_slope_bound in every dimension t<dt < dt<d (hypothesis hIH). Let U⊆CdU \subseteq \mathbb C^dU⊆Cd with n=dim⁡U<dn = \dim U < dn=dimU<d and ddim⁡(U∩S)≤ndim⁡Sd \dim(U \cap S) \le n \dim Sddim(U∩S)≤ndimS for every subspace S⊆CdS \subseteq \mathbb C^dS⊆Cd rational over Q\mathbb QQ, and let z1,…,zℓ∈U∩(Q+L)dz_1, \dots, z_\ell \in U \cap (\mathbb Q + \mathcal L)^dz1​,…,zℓ​∈U∩(Q+L)d be linearly independent over Q\mathbb QQ. Then (d−n) ℓ≤d n(d - n)\,\ell \le d\,n(d−n)ℓ≤dn.

The multiplicity estimate is the explicit hypothesis hME : MultiplicityEstimateElem (proved on the platform as DiazModulus.multiplicityEstimateElem).

Preamble
import Definitions.Def_DiazModulus
import Definitions.Def_MultiplicityEstimateElem
Formal statement
open Complex ComplexConjugate

namespace DiazModulus

theorem roy_rat_semistable_node (hME : MultiplicityEstimateElem) {d : ℕ}
    (hIH : ∀ t < d, ∀ (U' : Submodule ℂ (Fin t → ℂ)) (a b ℓ' : ℕ) (z' : Fin ℓ' → Fin t → ℂ),
        a < b → (∀ S : Submodule ℂ (Fin t → ℂ),
          Submodule.span ℂ {v : Fin t → ℂ | v ∈ S ∧ ∀ i, ∃ q : ℚ, v i = (q : ℂ)} = S →
          b * Module.finrank ℂ ↥(U' ⊓ S) ≤ a * Module.finrank ℂ ↥S) →
        (∀ i, z' i ∈ U') → (∀ i j, z' i j ∈ Submodule.span ℚ (insert (1 : ℂ) LogAlg)) →
        LinearIndependent ℚ z' → (b - a) * ℓ' ≤ b * Module.finrank ℂ ↥U')
    (U : Submodule ℂ (Fin d → ℂ)) (hUd : Module.finrank ℂ ↥U < d)
    (hss : ∀ S : Submodule ℂ (Fin d → ℂ),
      Submodule.span ℂ {v : Fin d → ℂ | v ∈ S ∧ ∀ i, ∃ q : ℚ, v i = (q : ℂ)} = S →
      d * Module.finrank ℂ ↥(U ⊓ S) ≤ Module.finrank ℂ ↥U * Module.finrank ℂ ↥S)
    {ℓ : ℕ} (z : Fin ℓ → Fin d → ℂ) (hzU : ∀ i, z i ∈ U)
    (hz : ∀ i j, z i j ∈ Submodule.span ℚ (insert (1 : ℂ) LogAlg)) (hzind : LinearIndependent ℚ z) :
    (d - Module.finrank ℂ ↥U) * ℓ ≤ d * Module.finrank ℂ ↥U := by
  sorry

end DiazModulus
Source
D. Roy, Matrices whose coefficients are linear forms in logarithms, J. Number Theory 41 (1992), 22–47, Th. 4 (p. 34), semistable case, over F = ℚ (Remark (i), p. 37, with ℚ + ℒ in place of ℒ̃); M. Waldschmidt, Diophantine Approximation on Linear Algebraic Groups, Grundlehren der mathematischen Wissenschaften 326, Springer, 2000, Prop. 11.19 (p. 402) for (ℚ, ℚ + ℒ) in the same case. The route is that of DiazModulus.roy_semistable_node, over ℚ and with m = 1: DALAG Th. 11.21 (DiazModulus.linear_subgroup_transcendence) is applied once, and the periods are counted through one extra rational subspace. Formal proof: Diaz modulus mission, 14 November 2026 (C. Perassi).

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