Linear-forms condition with arbitrary finite row and variable sets
ProvedGreenTao.linear_forms_finite_indicesadditive-combinatoricsgowers-uniformitygreen-tao
Let satisfy the linear-forms condition. Let be finite sets with and . Suppose rational coefficients have numerator and denominator bounded in absolute value by , each row is nonzero, and distinct rows are not rational multiples. For every sequence of translations ,
Rational coefficients are interpreted using the project's existing modular coefficient map. This is the existing linear-forms condition with arbitrary finite index sets, allowing direct use with cube vertices, faces, and disjoint unions of configurations.
Formalization Note. The finite variable set carries decidable equality so that Lean can construct its finite function space.
Preamble
import Definitions.Def_GreenTao_Pseudorandom import Mathlib open scoped BigOperators Topology open Filter GreenTao
Formal statement
theorem GreenTao.linear_forms_finite_indices (M : ℕ → ℕ+) (ν : GreenTao.Family M) (m₀ t₀ L₀ : ℕ)
(hν : GreenTao.LinearFormsCondition M ν m₀ t₀ L₀)
(I J : Type) [Fintype I] [Fintype J] [DecidableEq J]
(hm : Fintype.card I ≤ m₀) (ht : Fintype.card J ≤ t₀)
(L : I → J → ℚ)
(hL : ∀ i j, (L i j).num.natAbs ≤ L₀ ∧ (L i j).den ≤ L₀)
(hnonzero : ∀ i, ∃ j, L i j ≠ 0)
(hdistinct : ∀ i i', i ≠ i' → ¬ ∃ c : ℚ, ∀ j, L i j = c * L i' j)
(b : ∀ n, I → ZMod (M n : ℕ)) :
Tendsto (fun n => avg fun x : J → ZMod (M n : ℕ) =>
∏ i : I, ν n ((∑ j : J, ratCoeff (M n) (L i j) * x j) + b n i)) atTop (𝓝 1) := by sorrySource
Green and Tao, The primes contain arbitrarily long arithmetic progressions, https://arxiv.org/html/math/0404188v6, §3 Definition 3.1, equation (3.1). This theorem reindexes the project’s existing Fin-based condition using finite equivalences; it does not strengthen its hypotheses or bounds.