AFTD Auto-Formalizing Theoretical Domains

four_point_positive_kernel_primitive_boundary_decomposition

theoremverified

For two integer rows on four coordinates with independent first two columns and a strictly positive integer common kernel vector x, there are two nonzero nonnegative primitive integer common kernel vectors u,v, each having a zero coordinate, and positive integer D,c,d (represented in Z) such that D*x=c*u+d*v. This constructs the primitive boundary decomposition but does not yet prove their minimal-zero-sum interpretation or the box three-support length bound.

Statement

theorem four_point_positive_kernel_primitive_boundary_decomposition
    (a b x : Fin 4 → ℤ) (hx : ∀ i, 0 < x i)
    (hxa : (∑ i, x i*a i) = 0) (hxb : (∑ i, x i*b i) = 0)
    (hind : a 0*b 1 - a 1*b 0 ≠ 0) :
    ∃ (u v : Fin 4 → ℤ) (i j : Fin 4) (D c d : ℤ),
      (∀ k, 0 ≤ u k) ∧ (∀ k, 0 ≤ v k) ∧ u ≠ 0 ∧ v ≠ 0 ∧
      u i = 0 ∧ v j = 0 ∧ 0 < D ∧ 0 < c ∧ 0 < d ∧
      (∀ k, D*x k = c*u k + d*v k) ∧
      (∑ k, u k*a k) = 0 ∧ (∑ k, u k*b k) = 0 ∧
      (∑ k, v k*a k) = 0 ∧ (∑ k, v k*b k) = 0 ∧
      (∀ e : ℕ, (∀ k, (e : ℤ) ∣ u k) → e ∣ 1) ∧
      (∀ e : ℕ, (∀ k, (e : ℤ) ∣ v k) → e ∣ 1)
Source
Supporting lemma or explicit route obstruction for OP-197 / arXiv:2610.16562 Conjecture1, deep session2. The abstract two-ray fact and finite obstruction are not claimed novel or research-level; no part is the full box conjecture. The five-point kernel has columns (1,1),(1,0),(0,1),(-2,0),(0,-2).
Verified
28 Oct 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 4

The proof, explained

The proof begins by applying an existing boundary decomposition theorem to obtain nonnegative integer kernel vectors u₀ and v₀ (each vanishing at one coordinate) expressing D x = c u₀ + d v₀ with positive weights D, c, d. It then factors u₀ = g u and v₀ = h v into positive scalar contents g, h and primitive integer vectors u, v. Absorbing these scalar factors into the coefficients as c' = c g and d' = d h yields the primitive representation D x = c' u + d' v, while the vanishing coordinates and kernel relations are preserved because g and h are strictly positive.

  1. There exist nonzero nonnegative integer vectors u₀ and v₀ in the common kernel of a and b, each having at least one zero coordinate, decomposing D x = c u₀ + d v₀ for positive integers D, c, d.

    We invoke four_point_positive_kernel_boundary_decomposition, which establishes the boundary decomposition of x into two nonnegative kernel vectors u₀ and v₀, providing indices i and j where u₀(i) = 0 and v₀(j) = 0, along with positive scalars D, c, d.

    lines 17–18

    uses four_point_positive_kernel_boundary_decomposition

      obtain ⟨u0,v0,i,j,D,c,d,hu0,hv0,hun0,hvn0,hui0,hvj0,hD,hc,hd,hrel,hua0,hub0,hva0,hvb0⟩ :=
        four_point_positive_kernel_boundary_decomposition a b x hx hxa hxb hind
  2. The vectors u₀ and v₀ factor as u₀ = g u and v₀ = h v, where g, h are strictly positive integers and u, v are primitive nonnegative integer vectors.

    We apply nonnegative_integer_vector_primitive_factorization to u₀ and v₀, extracting their integer content as positive integers g and h, leaving nonzero nonnegative primitive vectors u and v. The positivity of g and h is then cast from ℕ to ℤ.

    lines 19–22

    uses nonnegative_integer_vector_primitive_factorization

      obtain ⟨g,u,hg,hu,hun,hgu,hpu⟩ := nonnegative_integer_vector_primitive_factorization u0 hu0 hun0
      obtain ⟨h,v,hh,hv,hvn,hhv,hpv⟩ := nonnegative_integer_vector_primitive_factorization v0 hv0 hvn0
      have hgz : (0 : ℤ)<g := by exact_mod_cast hg
      have hhz : (0 : ℤ)<h := by exact_mod_cast hh
  3. Scaling an integer vector by a strictly positive integer preserves membership in the kernel of any linear form.

    We prove the auxiliary lemma hker: if z₀ = t z with t > 0 and ∑ₖ z₀(k) r(k) = 0, then t ∑ₖ z(k) r(k) = ∑ₖ z₀(k) r(k) = 0 by distributing multiplication over the sum via Finset.mul_sum and Finset.sum_congr. Since t ≠ 0, mul_eq_zero.mp implies ∑ₖ z(k) r(k) = 0.

    lines 23–33

    uses Finset.mul_sum, Finset.sum_congr, mul_eq_zero.mp

      have hker (z0 z : Fin 4 → ℤ) (t : ℕ) (ht : 0<t)
          (hfac : ∀ k, z0 k=(t : ℤ)*z k) (r : Fin 4 → ℤ)
          (hz0 : (∑ k, z0 k*r k)=0) : (∑ k, z k*r k)=0 := by
        have he : (t : ℤ)*(∑ k,z k*r k)=0 := by
          calc
            _ = ∑ k,z0 k*r k := by
              rw [Finset.mul_sum]
              apply Finset.sum_congr rfl
              intro k hk; rw [hfac k]; ring
            _ = 0 := hz0
        exact (mul_eq_zero.mp he).resolve_left (by exact_mod_cast ht.ne')
  4. The primitive vectors u and v preserve the zero coordinates: u(i) = 0 and v(j) = 0.

    Since u₀(i) = g u(i) = 0 and g > 0 (hence g ≠ 0), mul_eq_zero.mp forces u(i) = 0. Symmetrically, v₀(j) = h v(j) = 0 with h > 0 implies v(j) = 0.

    lines 34–41

    uses mul_eq_zero.mp

      have hui : u i=0 := by
        have he := hgu i
        rw [hui0] at he
        exact (mul_eq_zero.mp he.symm).resolve_left hgz.ne'
      have hvj : v j=0 := by
        have he := hhv j
        rw [hvj0] at he
        exact (mul_eq_zero.mp he.symm).resolve_left hhz.ne'
  5. The primitive vectors u, v, indices i, j, scalar D, and scaled weights c' = c g and d' = d h satisfy all conditions of the theorem except the linear combination identity.

    We instantiate the existential witnesses with u, v, i, j, D, c * g, and d * h. The products c * g and d * h are positive by mul_pos since c, d, g, h > 0. The kernel conditions with respect to a and b are discharged using the helper lemma hker on u₀ = g u and v₀ = h v, and primitivity holds by the factorization.

    lines 42–45

    uses mul_pos

      refine ⟨u,v,i,j,D,c*(g : ℤ),d*(h : ℤ),hu,hv,hun,hvn,hui,hvj,hD,
        mul_pos hc hgz,mul_pos hd hhz,?_,
        hker u0 u g hg hgu a hua0,hker u0 u g hg hgu b hub0,
        hker v0 v h hh hhv a hva0,hker v0 v h hh hhv b hvb0,hpu,hpv⟩
  6. The linear combination identity D x(k) = (c g) u(k) + (d h) v(k) holds for every coordinate k.

    Substituting u₀(k) = g u(k) and v₀(k) = h v(k) into D x(k) = c u₀(k) + d v₀(k) yields c (g u(k)) + d (h v(k)) = (c g) u(k) + (d h) v(k), which holds by associativity and distributivity of multiplication in ℤ.

    lines 46–48

      intro k
      rw [hrel k,hgu k,hhv k]
      ring
four_point_positive_kernel_boundary_decomposition
Under linear independence and kernel assumptions, a strictly positive integer vector in ℝ⁴ admits a conical decomposition Dx = c u₀ + d v₀ into nonzero nonnegative kernel vectors u₀, v₀ with positive weights D, c, d, such that each vector has at least one zero coordinate.
nonnegative_integer_vector_primitive_factorization
Any nonzero nonnegative integer vector factors as a strictly positive integer (the greatest common divisor of its entries) times a primitive integer vector.
Finset.mul_sum
Multiplication by a scalar distributes over a finite sum: a ∑ᵢ f(i) = ∑ᵢ a f(i).
Finset.sum_congr
Two finite sums over the same index set are equal if their summands agree at every index.
mul_eq_zero.mp
If a product of two integers is zero, then at least one of the factors is zero.
mul_pos
The product of two strictly positive numbers is strictly positive.

Written by a model (gemini-3.8-flash) from the Lean proof and the goal states Lean reported at each step. Lean verified the proof; this explanation is a reading aid and is not itself machine-checked.

Lean source view module on GitHub

open Finsetset_option maxHeartbeats 1000000set_option maxRecDepth 4000theorem four_point_positive_kernel_primitive_boundary_decomposition    (a b x : Fin 4 → ℤ) (hx : ∀ i, 0 < x i)    (hxa : (∑ i, x i*a i) = 0) (hxb : (∑ i, x i*b i) = 0)    (hind : a 0*b 1 - a 1*b 0 ≠ 0) :    ∃ (u v : Fin 4 → ℤ) (i j : Fin 4) (D c d : ℤ),      (∀ k, 0 ≤ u k) ∧ (∀ k, 0 ≤ v k) ∧ u ≠ 0 ∧ v ≠ 0 ∧      u i = 0 ∧ v j = 0 ∧ 0 < D ∧ 0 < c ∧ 0 < d ∧      (∀ k, D*x k = c*u k + d*v k) ∧      (∑ k, u k*a k) = 0 ∧ (∑ k, u k*b k) = 0 ∧      (∑ k, v k*a k) = 0 ∧ (∑ k, v k*b k) = 0 ∧      (∀ e : ℕ, (∀ k, (e : ℤ) ∣ u k) → e ∣ 1) ∧      (∀ e : ℕ, (∀ k, (e : ℤ) ∣ v k) → e ∣ 1) := by  obtain ⟨u0,v0,i,j,D,c,d,hu0,hv0,hun0,hvn0,hui0,hvj0,hD,hc,hd,hrel,hua0,hub0,hva0,hvb0⟩ :=    four_point_positive_kernel_boundary_decomposition a b x hx hxa hxb hind  obtain ⟨g,u,hg,hu,hun,hgu,hpu⟩ := nonnegative_integer_vector_primitive_factorization u0 hu0 hun0  obtain ⟨h,v,hh,hv,hvn,hhv,hpv⟩ := nonnegative_integer_vector_primitive_factorization v0 hv0 hvn0  have hgz : (0 : ℤ)<g := by exact_mod_cast hg  have hhz : (0 : ℤ)<h := by exact_mod_cast hh  have hker (z0 z : Fin 4 → ℤ) (t : ℕ) (ht : 0<t)      (hfac : ∀ k, z0 k=(t : ℤ)*z k) (r : Fin 4 → ℤ)      (hz0 : (∑ k, z0 k*r k)=0) : (∑ k, z k*r k)=0 := by    have he : (t : ℤ)*(∑ k,z k*r k)=0 := by      calc        _ = ∑ k,z0 k*r k := by          rw [Finset.mul_sum]          apply Finset.sum_congr rfl          intro k hk; rw [hfac k]; ring        _ = 0 := hz0    exact (mul_eq_zero.mp he).resolve_left (by exact_mod_cast ht.ne')  have hui : u i=0 := by    have he := hgu i    rw [hui0] at he    exact (mul_eq_zero.mp he.symm).resolve_left hgz.ne'  have hvj : v j=0 := by    have he := hhv j    rw [hvj0] at he    exact (mul_eq_zero.mp he.symm).resolve_left hhz.ne'  refine ⟨u,v,i,j,D,c*(g : ℤ),d*(h : ℤ),hu,hv,hun,hvn,hui,hvj,hD,    mul_pos hc hgz,mul_pos hd hhz,?_,    hker u0 u g hg hgu a hua0,hker u0 u g hg hgu b hub0,    hker v0 v h hh hhv a hva0,hker v0 v h hh hhv b hvb0,hpu,hpv⟩  intro k  rw [hrel k,hgu k,hhv k]  ring
See it in the graphDiscuss this resultChallenge it