AFTD Auto-Formalizing Theoretical Domains

mordell_ten_suboptimal

theoremverifiedformalization

In Mordell's problem for ten points in the complex plane, the supremum sSup of { d : ℝ | ∃ z : Fin 10 → ℂ, (∑ i : Fin 10, ‖z i‖^2 = 10) ∧ d = ∏ i : Fin 10, ∏ j : Fin 10 with i < j, ‖z i - z j‖^2 } strictly exceeds (10 : ℝ)^45 / (9 : ℝ)^36.

Statement

theorem mordell_ten_suboptimal :
    sSup { d : ℝ | ∃ z : Fin 10 → ℂ, (∑ i : Fin 10, ‖z i‖ ^ 2 = 10) ∧
      d = ∏ i : Fin 10, ∏ j : Fin 10 with i < j, ‖z i - z j‖ ^ 2 } >
    (10 : ℝ) ^ 45 / (9 : ℝ) ^ 36
Formalizes
arXiv:2611.03088 Proposition 5.1
Verified
05 Nov 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 1
  • mordell_ten_exists_feasible_gt_chalkThere exists a configuration of ten points z : Fin 10 → ℂ in the complex plane satisfying ∑ i : Fin 10, ‖z i‖^2 = 10 such that ∏ i : Fin…

Read back from the Lean

The supremum (in ℝ) of the set of all real numbers d for which there exists a function z : Fin 10 → ℂ such that ∑ i : Fin 10, ‖z i‖^2 = 10 and d = ∏ i : Fin 10, ∏ (j : Fin 10 with i < j), ‖z i - z j‖^2 is strictly greater than (10 : ℝ)^45 / (9 : ℝ)^36.

Written by a model that saw only the Lean, never the English above. If the two disagree, that is worth a challenge.

The proof, explained

The theorem proves that in Mordell's problem for ten points in the complex plane, the supremum of the product of pairwise squared distances strictly exceeds 10^45 / 9^36. The proof first shows in an auxiliary lemma that the set of achievable products is bounded above by establishing that each pairwise squared distance cannot exceed 40. It then imports a feasible 10-point witness configuration whose product strictly exceeds 10^45 / 9^36, and concludes using the order-theoretic supremum property (lt_csSup_of_lt) that the supremum of the set strictly exceeds this threshold.

  1. To show that the set of products is bounded above, it suffices to show that the product for any feasible configuration is bounded by ∏ i < j, 40.

    We select ∏ i < j, 40 (which equals 40^45) as a candidate upper bound. Taking an arbitrary value d in the set, we unpack the witness configuration z : Fin 10 → ℂ satisfying ∑ i, ‖z i‖^2 = 10, reducing the goal to bounding its product of pairwise squared distances by ∏ i < j, 40.

    lines 4–5

      use ∏ i : Fin 10, ∏ j : Fin 10 with i < j, (40 : ℝ)
      rintro d ⟨z, hz, rfl⟩
  2. It suffices to show that each pairwise squared distance satisfies ‖z i - z j‖^2 ≤ 40 for all i < j.

    By monotonicity of finite products (Finset.prod_le_prod), because each squared distance factor is non-negative (sq_nonneg, Finset.prod_nonneg), bounding the full product reduces to showing that each individual factor satisfies ‖z i - z j‖^2 ≤ 40.

    lines 6–13

    uses Finset.prod_le_prod, Finset.prod_nonneg, sq_nonneg

      apply Finset.prod_le_prod
      · intro i _
        exact Finset.prod_nonneg fun _ _ => sq_nonneg _
      · intro i _
        apply Finset.prod_le_prod
        · intro j _
          exact sq_nonneg _
        · intro j _
  3. Each pairwise squared distance is at most 40, completing the proof that the set is bounded above.

    By the triangle inequality (norm_sub_le), ‖z i - z j‖ ≤ ‖z i‖ + ‖z j‖, so squaring both sides gives ‖z i - z j‖^2 ≤ (‖z i‖ + ‖z j‖)^2 using non-negativity of norms (norm_nonneg). Expanding algebraically gives (‖z i‖ + ‖z j‖)^2 = 2(‖z i‖^2 + ‖z j‖^2) - (‖z i‖ - ‖z j‖)^2 ≤ 2(‖z i‖^2 + ‖z j‖^2) since squares are non-negative (sq_nonneg). Because the sum of all ten squared norms is 10 and each term is non-negative (sq_nonneg), each individual term satisfies ‖z i‖^2 ≤ 10 and ‖z j‖^2 ≤ 10 (Finset.single_le_sum, Finset.mem_univ). Linear arithmetic then deduces ‖z i - z j‖^2 ≤ 2(10 + 10) = 40.

    lines 14–25

    uses norm_sub_le, norm_nonneg, sq_nonneg, Finset.single_le_sum, Finset.mem_univ

          have h1 : ‖z i - z j‖ ≤ ‖z i‖ + ‖z j‖ := norm_sub_le (z i) (z j)
          have h2 : ‖z i - z j‖ ^ 2 ≤ (‖z i‖ + ‖z j‖) ^ 2 := by
            nlinarith [norm_nonneg (z i - z j), norm_nonneg (z i), norm_nonneg (z j)]
          have h3 : 2 * (‖z i‖ ^ 2 + ‖z j‖ ^ 2) - (‖z i‖ + ‖z j‖) ^ 2 = (‖z i‖ - ‖z j‖) ^ 2 := by ring
          have h4 : (‖z i‖ - ‖z j‖) ^ 2 ≥ 0 := sq_nonneg _
          have h_zi : ‖z i‖ ^ 2 ≤ 10 := by
            have := Finset.single_le_sum (fun (k : Fin 10) _ => sq_nonneg (‖z k‖)) (Finset.mem_univ i)
            linarith
          have h_zj : ‖z j‖ ^ 2 ≤ 10 := by
            have := Finset.single_le_sum (fun (k : Fin 10) _ => sq_nonneg (‖z k‖)) (Finset.mem_univ j)
            linarith
          linarith
  4. There exists a feasible configuration whose squared-distance product belongs to S and strictly exceeds 10^45 / 9^36.

    We instantiate the witness configuration z from mordell_ten_exists_feasible_gt_chalk, which satisfies the constraint ∑ i, ‖z i‖^2 = 10 and whose pairwise squared-distance product strictly exceeds 10^45 / 9^36. By definition of the set S, this product is an element of S.

    lines 32–36

    uses mordell_ten_exists_feasible_gt_chalk

      obtain ⟨z, hz_sum, hz_gt⟩ := mordell_ten_exists_feasible_gt_chalk
      let S := { d : ℝ | ∃ z : Fin 10 → ℂ, (∑ i : Fin 10, ‖z i‖ ^ 2 = 10) ∧
        d = ∏ i : Fin 10, ∏ j : Fin 10 with i < j, ‖z i - z j‖ ^ 2 }
      have hmem : (∏ i : Fin 10, ∏ j : Fin 10 with i < j, ‖z i - z j‖ ^ 2) ∈ S :=
        ⟨z, hz_sum, rfl⟩
  5. The supremum of S strictly exceeds 10^45 / 9^36.

    By lt_csSup_of_lt, if a set S ⊆ ℝ is bounded above (established by mordell_ten_suboptimal_bddAbove) and contains an element strictly greater than a threshold c, then sSup S > c. Applying this to our witness configuration product completes the proof.

    line 37

    uses lt_csSup_of_lt, mordell_ten_suboptimal_bddAbove

      exact lt_csSup_of_lt mordell_ten_suboptimal_bddAbove hmem hz_gt
Finset.prod_le_prod
If 0 ≤ f(x) ≤ g(x) for all x in a finite set s, then ∏ x ∈ s, f(x) ≤ ∏ x ∈ s, g(x).
Finset.prod_nonneg
A finite product of non-negative elements is non-negative.
sq_nonneg
The square of any element in an ordered semiring is non-negative (a^2 ≥ 0).
norm_sub_le
The norm of a difference is bounded by the sum of the norms: ‖a - b‖ ≤ ‖a‖ + ‖b‖.
norm_nonneg
The norm of any vector in a seminormed group is non-negative (‖a‖ ≥ 0).
Finset.single_le_sum
In a finite sum of non-negative terms, each individual term is at most the total sum.
Finset.mem_univ
Every element of a finite type belongs to the universal finset Finset.univ.
mordell_ten_exists_feasible_gt_chalk
There exists a 10-point configuration in ℂ with ∑ i, ‖z i‖^2 = 10 whose pairwise squared-distance product strictly exceeds 10^45 / 9^36.
lt_csSup_of_lt
In a conditionally complete lattice, if a set s is bounded above and contains an element strictly greater than b, then b < sSup s.
mordell_ten_suboptimal_bddAbove
The set of pairwise squared-distance products under the constraint ∑ i, ‖z i‖^2 = 10 is bounded above in ℝ.

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

lemma mordell_ten_suboptimal_bddAbove :    BddAbove { d : ℝ | ∃ z : Fin 10 → ℂ, (∑ i : Fin 10, ‖z i‖ ^ 2 = 10) ∧      d = ∏ i : Fin 10, ∏ j : Fin 10 with i < j, ‖z i - z j‖ ^ 2 } := by  use ∏ i : Fin 10, ∏ j : Fin 10 with i < j, (40 : ℝ)  rintro d ⟨z, hz, rfl⟩  apply Finset.prod_le_prod  · intro i _    exact Finset.prod_nonneg fun _ _ => sq_nonneg _  · intro i _    apply Finset.prod_le_prod    · intro j _      exact sq_nonneg _    · intro j _      have h1 : ‖z i - z j‖ ≤ ‖z i‖ + ‖z j‖ := norm_sub_le (z i) (z j)      have h2 : ‖z i - z j‖ ^ 2 ≤ (‖z i‖ + ‖z j‖) ^ 2 := by        nlinarith [norm_nonneg (z i - z j), norm_nonneg (z i), norm_nonneg (z j)]      have h3 : 2 * (‖z i‖ ^ 2 + ‖z j‖ ^ 2) - (‖z i‖ + ‖z j‖) ^ 2 = (‖z i‖ - ‖z j‖) ^ 2 := by ring      have h4 : (‖z i‖ - ‖z j‖) ^ 2 ≥ 0 := sq_nonneg _      have h_zi : ‖z i‖ ^ 2 ≤ 10 := by        have := Finset.single_le_sum (fun (k : Fin 10) _ => sq_nonneg (‖z k‖)) (Finset.mem_univ i)        linarith      have h_zj : ‖z j‖ ^ 2 ≤ 10 := by        have := Finset.single_le_sum (fun (k : Fin 10) _ => sq_nonneg (‖z k‖)) (Finset.mem_univ j)        linarith      linarith/-- Suboptimality of the centered regular polygon configuration for ten points in Mordell's problem. -/theorem mordell_ten_suboptimal :    sSup { d : ℝ | ∃ z : Fin 10 → ℂ, (∑ i : Fin 10, ‖z i‖ ^ 2 = 10) ∧      d = ∏ i : Fin 10, ∏ j : Fin 10 with i < j, ‖z i - z j‖ ^ 2 } >    (10 : ℝ) ^ 45 / (9 : ℝ) ^ 36 := by  obtain ⟨z, hz_sum, hz_gt⟩ := mordell_ten_exists_feasible_gt_chalk  let S := { d : ℝ | ∃ z : Fin 10 → ℂ, (∑ i : Fin 10, ‖z i‖ ^ 2 = 10) ∧    d = ∏ i : Fin 10, ∏ j : Fin 10 with i < j, ‖z i - z j‖ ^ 2 }  have hmem : (∏ i : Fin 10, ∏ j : Fin 10 with i < j, ‖z i - z j‖ ^ 2) ∈ S :=    ⟨z, hz_sum, rfl⟩  exact lt_csSup_of_lt mordell_ten_suboptimal_bddAbove hmem hz_gt
See it in the graphDiscuss this resultChallenge it