Lemma 9: nontrivial column combinations have few zeros
ProvedQAlgorithms.Papers.Independent.Intake100.P43.column_combination_few_zerosLet , let be totally nonsingular, and select distinct columns using an injection . For coefficients that are not all zero, put . Then
Some individual coefficients may vanish. This corrected nontrivial-combination form is the one required by the source’s proof and construction.
import Definitions.Def_QAlgorithms_Papers_Independent_Intake100_P43
noncomputable section
open scoped BigOperators Matrix ComplexConjugate
namespace QAlgorithms.Papers.Independent.Intake100.P43
open QAlgorithms.Papers.Independent.Dependency50.P50 (Bipartite schmidtNumber)
open QAlgorithms.Papers.Independent.Information100.P03 (PureEnsemble)
/-- PDF pp.2–3, Lemma 9, corrected from its proof: printed "any linear
combination" must mean a NONTRIVIAL combination of DISTINCT selected columns.
The all-zero coefficient vector contradicts the literal printed statement.
No requirement that all coefficients be individually nonzero is imposed. -/
theorem column_combination_few_zeros (m n : ℕ) (hn : 1 ≤ n) (hnm : n ≤ m)
(M : Matrix (Fin m) (Fin m) ℂ) (hM : TotallyNonsingular M)
(cols : Fin n ↪ Fin m) (c : Fin n → ℂ) (hc : c ≠ 0) :
(Finset.univ.filter (fun i : Fin m =>
(∑ j : Fin n, c j * M i (cols j)) = 0)).card ≤ n - 1 := by sorry
end QAlgorithms.Papers.Independent.Intake100.P43
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For all natural numbers with , every complex matrix whose every square submatrix selected by injective row and column lists has nonzero determinant, every injection , and every coefficient vector with at least one nonzero coefficient, the number of indices satisfying is at most . The condition on square submatrices includes the empty submatrix, whose determinant is . The subtraction is natural-number subtraction, and ensures that it has its usual integer value.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.