AFTD Auto-Formalizing Theoretical Domains

apn_family_fa_component_eq_cubic

theoremverified

Let a and β be elements of a finite field E of characteristic 2 such that the absolute trace of β * a⁻¹ is 0. Then for all x in E, the trace of β * F_a(x) equals the trace of β * x^3.

Statement

theorem apn_family_fa_component_eq_cubic {E : Type*} [Field E] [Fintype E] [Algebra (ZMod 2) E] (a β : E) (hH : Algebra.trace (ZMod 2) E (β * a⁻¹) = 0) (x : E) : Algebra.trace (ZMod 2) E (β * apn_family_fa E a x) = Algebra.trace (ZMod 2) E (β * x ^ 3)
Source
arXiv:2611.10269 Sec. 7
Verified
11 Nov 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 1
  • apn_family_faFor an element a in a finite field E of characteristic 2, the function F_a(x) is defined by x^3 + a⁻¹ * Tr(a^3 * x^9).
Used by 1
  • spread_line_contained_in_non_bentLet E be a finite field equipped with an algebra structure over ZMod 2 such that the cardinality of E is 2^(2m), where m is an odd natural…

Read back from the Lean

Let $E$ be a finite field equipped with an algebra structure over $\mathbb{Z}/2\mathbb{Z}$. For any elements $a, \beta \in E$, if the algebra trace $\operatorname{Tr}_{E/(\mathbb{Z}/2\mathbb{Z})}(\beta \cdot a^{-1}) = 0$, then for all $x \in E$, the algebra trace $\operatorname{Tr}_{E/(\mathbb{Z}/2\mathbb{Z})}(\beta \cdot \operatorname{apn\_family\_fa}(E, a, x))$ is equal to $\operatorname{Tr}_{E/(\mathbb{Z}/2\mathbb{Z})}(\beta \cdot x^3)$.

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 map F_a(x) differs from x^3 by a term proportional to a⁻¹, scaled by the base field element Tr(a^3 * x^9). By linearity of the trace, Tr(β * F_a(x)) decomposes into Tr(β * x^3) + Tr(a^3 * x^9) * Tr(β * a⁻¹). Because β is assumed to lie on the hyperplane orthogonal to a⁻¹ under the trace pairing (Tr(β * a⁻¹) = 0), this correction term vanishes, leaving Tr(β * x^3).

  1. Tr(β * F_a(x)) expands to Tr(β * x^3) + Tr(algebraMap(Tr(a^3 * x^9)) * (β * a⁻¹)).

    Unfolding the definition of F_a(x) = x^3 + a⁻¹ * algebraMap(Tr(a^3 * x^9)), we distribute β across the sum using mul_add and commutative ring arithmetic to isolate the base-field scalar (hypothesis h1). Applying additivity of the trace via map_add splits the trace into the sum of the traces of the two terms.

    lines 3–7

    uses mul_add, map_add

      unfold apn_family_fa
      have h1 : β * (x ^ 3 + a⁻¹ * algebraMap (ZMod 2) E (Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9)))
          = β * x ^ 3 + algebraMap (ZMod 2) E (Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9)) * (β * a⁻¹) := by
        rw [mul_add]; ring
      rw [h1, map_add]
  2. The correction term Tr(algebraMap(Tr(a^3 * x^9)) * (β * a⁻¹)) is zero.

    Using Algebra.smul_def, the product is rewritten as the scalar multiplication of β * a⁻¹ by the scalar c = Tr(a^3 * x^9) ∈ ℤ/2ℤ. Since the trace is a linear map over ℤ/2ℤ, LinearMap.map_smul pulls this scalar out, giving c * Tr(β * a⁻¹) (sub-claim hc). Because Tr(β * a⁻¹) = 0 by hypothesis hH, this product is zero by mul_zero.

    lines 8–12

    uses Algebra.smul_def, LinearMap.map_smul, mul_zero

      have h2 : Algebra.trace (ZMod 2) E (algebraMap (ZMod 2) E (Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9)) * (β * a⁻¹)) = 0 := by
        have hc : Algebra.trace (ZMod 2) E (algebraMap (ZMod 2) E (Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9)) * (β * a⁻¹))
            = Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9) * Algebra.trace (ZMod 2) E (β * a⁻¹) := by
          rw [← Algebra.smul_def, LinearMap.map_smul]; rfl
        rw [hc, hH, mul_zero]
  3. Tr(β * x^3) + 0 simplifies to Tr(β * x^3), completing the proof.

    Substituting the vanishing of the correction term (h2) into the main goal leaves Tr(β * x^3) + 0 = Tr(β * x^3), which is simplified by add_zero.

    line 13

    uses add_zero

      rw [h2, add_zero]
mul_add
Left distributivity of multiplication over addition: a * (b + c) = a * b + a * c.
map_add
An additive homomorphism preserves addition: f(x + y) = f x + f y.
Algebra.smul_def
Scalar multiplication by a base ring element equals multiplication by its image under the algebra map: r • x = algebraMap R A r * x.
LinearMap.map_smul
A linear map preserves scalar multiplication: f(c • x) = c • f x.
mul_zero
Multiplying any element by zero on the right gives zero: a * 0 = 0.
add_zero
Adding zero on the right leaves an element unchanged: a + 0 = a.

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

/-- On the hyperplane of elements orthogonal to a⁻¹, the component of F_a coincides with that of x^3. -/theorem apn_family_fa_component_eq_cubic {E : Type*} [Field E] [Fintype E] [Algebra (ZMod 2) E] (a β : E) (hH : Algebra.trace (ZMod 2) E (β * a⁻¹) = 0) (x : E) : Algebra.trace (ZMod 2) E (β * apn_family_fa E a x) = Algebra.trace (ZMod 2) E (β * x ^ 3) := by  unfold apn_family_fa  have h1 : β * (x ^ 3 + a⁻¹ * algebraMap (ZMod 2) E (Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9)))      = β * x ^ 3 + algebraMap (ZMod 2) E (Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9)) * (β * a⁻¹) := by    rw [mul_add]; ring  rw [h1, map_add]  have h2 : Algebra.trace (ZMod 2) E (algebraMap (ZMod 2) E (Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9)) * (β * a⁻¹)) = 0 := by    have hc : Algebra.trace (ZMod 2) E (algebraMap (ZMod 2) E (Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9)) * (β * a⁻¹))        = Algebra.trace (ZMod 2) E (a ^ 3 * x ^ 9) * Algebra.trace (ZMod 2) E (β * a⁻¹) := by      rw [← Algebra.smul_def, LinearMap.map_smul]; rfl    rw [hc, hH, mul_zero]  rw [h2, add_zero]
See it in the graphDiscuss this resultChallenge it