Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Genericity certificate: the other rows have the full set of normalized common zeros

Proved
HararySachs.Hyper.generic_row_points

by milia · Nov 13, 2026 · Mathlib 0df444a (Lean v4.33.1)

hypergraphsresultantsspectral-graph-theory

Let n≥1n \ge 1n≥1, k≥2k \ge 2k≥2, r=k−1r = k-1r=k−1, let Res⁡\operatorname{Res}Res be the normalized resultant of nnn forms of degree rrr, and fix a row iii. For a complex tensor aaa, let

Fja(x)=xjr−∑taj,t xt1⋯xtr(j=1,…,n).F_j^a(x) = x_j^{r} - \sum_t a_{j,t}\,x_{t_1}\cdots x_{t_r} \qquad (j = 1,\dots,n).Fja​(x)=xjr​−t∑​aj,t​xt1​​⋯xtr​​(j=1,…,n).

Then there is a nonzero polynomial Δ\DeltaΔ in the entries of aaa with the following property. Whenever Δ(a)≠0\Delta(a) \ne 0Δ(a)=0:

  1. there are (k−1)n−1(k-1)^{n-1}(k−1)n−1 distinct points p∈Cnp \in \mathbb{C}^np∈Cn with pi=1p_i = 1pi​=1 that are common zeros of FjaF_j^aFja​ for all j≠ij \ne ij=i;
  2. the resultant of the system obtained by replacing FiaF_i^aFia​ with xirx_i^{r}xir​ is nonzero.

Condition 2 says that the other equations have no common zero on the hyperplane xi=0x_i = 0xi​=0. Together with condition 1 it says that they are in general position: their common zeros are (k−1)n−1(k-1)^{n-1}(k−1)n−1 distinct points, all off that hyperplane. This is the hypothesis under which Poisson's product formula takes its simple form, and since Δ≠0\Delta \ne 0Δ=0 it holds on a Zariski-dense set.

Formalization Note The conclusion is stated with explicit forms X j ^ (k-1) - ∑ t, C (a (j,t)) * ∏ s, X (t s); the resultant in condition 2 is the evaluation of res at the coefficients of the system with row iii replaced by X i ^ (k-1).

Preamble
import Mathlib
import Definitions.Def_HararySachs_Hyper_CharPoly
Formal statement
namespace HararySachs.Hyper

theorem generic_row_points (n k : ℕ) (hk : 2 ≤ k) (hn : 1 ≤ n)
    (res : MvPolynomial (ResVar n (fun _ => k - 1)) ℤ)
    (hres : IsResultant n (fun _ => k - 1) res) (i : Fin n) :
    ∃ Δ : MvPolynomial (Fin n × (Fin (k - 1) → Fin n)) ℂ, Δ ≠ 0 ∧
      ∀ a : Fin n × (Fin (k - 1) → Fin n) → ℂ, MvPolynomial.eval a Δ ≠ 0 →
        (∃ P : Finset (Fin n → ℂ), P.card = (k - 1) ^ (n - 1) ∧
          ∀ p ∈ P, p i = 1 ∧ ∀ j, j ≠ i →
            MvPolynomial.eval p (MvPolynomial.X j ^ (k - 1) -
              ∑ t : Fin (k - 1) → Fin n,
                MvPolynomial.C (a (j, t)) * ∏ s, MvPolynomial.X (t s)) = 0) ∧
        MvPolynomial.aeval (fun v : ResVar n (fun _ => k - 1) =>
            MvPolynomial.coeff v.2.1 (Function.update
              (fun j => MvPolynomial.X j ^ (k - 1) -
                ∑ t : Fin (k - 1) → Fin n,
                  MvPolynomial.C (a (j, t)) * ∏ s, MvPolynomial.X (t s))
              i (MvPolynomial.X i ^ (k - 1)) v.1))
          (MvPolynomial.map (Int.castRingHom ℂ) res) ≠ 0 := by sorry

end HararySachs.Hyper
Source
Bezout count for n−1n-1n−1 forms of degree k−1k-1k−1 in nnn variables, used in the Poisson product formula of Morozov and Shakirov, Analogue of the identity Log Det = Trace Log for resultants, arXiv:0804.4632v3, Section 3 ("suppose that this system is in general position").

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