AFTD Auto-Formalizing Theoretical Domains

subfield_element_trace_zero

theoremverified

Let E be a finite field of characteristic 2 with 2^(2m) elements. Then the absolute trace of any element t in E satisfying t^(2^m) = t is zero in ZMod 2.

Statement

theorem subfield_element_trace_zero {E : Type*} [Field E] [Fintype E] [Algebra (ZMod 2) E] (m : ℕ) (hcard : Fintype.card E = 2 ^ (2 * m)) (t : E) (ht : t ^ (2 ^ m) = t) : Algebra.trace (ZMod 2) E t = 0
Source
arXiv:2611.10269 Sec. 7
Verified
11 Nov 2026
Axioms
Classical.choiceQuot.soundpropext
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 that is an algebra over $\mathbb{Z}/2\mathbb{Z}$. Let $m \in \mathbb{N}$ be such that the cardinality of $E$ is $2^{2m}$. Let $t \in E$ be an element satisfying $t^{2^m} = t$. Then the trace of $t$ from $E$ to $\mathbb{Z}/2\mathbb{Z}$ (i.e., $\operatorname{Tr}_{E / (\mathbb{Z}/2\mathbb{Z})}(t)$) equals $0$.

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 proof expresses the absolute trace of t over ZMod 2 as the sum of its 2m Galois conjugates, ∑ i < 2m, t^(2^i). Because t^(2^m) = t, this sum decomposes into two identical halves of m terms each. Since the ambient field has characteristic 2, adding any element to itself yields zero, and injectivity of the canonical embedding ZMod 2 ↪ E concludes that the trace is zero.

  1. The degree of the extension [E : ZMod 2] is 2m.

    The cardinality of a finite-dimensional vector space over a finite division ring is the base field cardinality raised to the dimension. Here, |E| = 2^(2m) and |ZMod 2| = 2, so 2^(2m) = 2^[E : ZMod 2]. Because x ↦ 2^x is injective on ℕ, the dimension [E : ZMod 2] must be 2m.

    lines 3–6

    uses Module.card_eq_pow_finrank, ZMod.card, Nat.pow_right_injective

      have hrank : Module.finrank (ZMod 2) E = 2 * m := by
        have h := Module.card_eq_pow_finrank (K := ZMod 2) (V := E)
        rw [ZMod.card, hcard] at h
        exact Nat.pow_right_injective (by norm_num) h.symm
  2. The image in E of the absolute trace of t equals ∑ i ∈ range(2m), t^(2^i).

    In a finite field extension, the trace of an element is the sum of its iterates under Frobenius, mapped into the field. Substituting the dimension 2m and the base field cardinality |ZMod 2| = 2 into this general trace formula expresses the image of the trace as the sum of t^(2^i) for i from 0 to 2m - 1.

    lines 7–9

    uses FiniteField.algebraMap_trace_eq_sum_pow

      have htrace := FiniteField.algebraMap_trace_eq_sum_pow (ZMod 2) E t
      have hcard2 : Nat.card (ZMod 2) = 2 := by simp
      rw [hrank, hcard2] at htrace
  3. The sum of 2m terms splits into two identical sums of m terms.

    Splitting the range of length 2m = m + m gives the sum of the first m terms, ∑ x ∈ range(m), t^(2^x), plus the sum of the remaining m terms, ∑ x ∈ range(m), t^(2^(m + x)). In each term of the second sum, the exponent is 2^(m + x) = (2^m) * (2^x), so t^(2^(m + x)) = (t^(2^m))^(2^x). Using the hypothesis that t^(2^m) = t, this simplifies to t^(2^x), showing that the two sums are term-by-term identical.

    lines 10–17

    uses two_mul, Finset.sum_range_add, Finset.sum_congr, pow_add, pow_mul

      have hsum : ∑ i ∈ Finset.range (2 * m), t ^ (2 ^ i) = 0 := by
        have h2m : 2 * m = m + m := two_mul m
        rw [h2m, Finset.sum_range_add]
        have h_eq : ∑ x ∈ Finset.range m, t ^ (2 ^ (m + x)) = ∑ x ∈ Finset.range m, t ^ (2 ^ x) := by
          apply Finset.sum_congr rfl
          intro x _
          rw [pow_add, pow_mul, ht]
        rw [h_eq]
  4. 2 = 0 in E.

    Because E is an algebra over ZMod 2, the element 2 in E is the image of 2 ∈ ZMod 2 under the algebra map. Since 2 = 0 in ZMod 2 and ring homomorphisms preserve 0, 2 is zero in E.

    lines 18–22

    uses map_ofNat, map_zero

        have h2 : (2 : E) = 0 := by
          have h : (2 : E) = algebraMap (ZMod 2) E (2 : ZMod 2) := (map_ofNat (algebraMap (ZMod 2) E) 2).symm
          rw [h]
          have : (2 : ZMod 2) = 0 := rfl
          rw [this, map_zero]
  5. Adding any element of E to itself gives 0, so the sum ∑ i ∈ range(2m), t^(2^i) equals 0.

    For any s in E, s + s = 2 * s = 0 * s = 0 since 2 = 0 in E. Applying this to the sum of the two identical halves shows that the entire sum vanishes.

    lines 23–28

    uses two_mul, zero_mul

        have h_add : ∀ s : E, s + s = 0 := by
          intro s
          calc s + s = 2 * s := (two_mul s).symm
          _ = 0 * s := by rw [h2]
          _ = 0 := zero_mul s
        exact h_add _
  6. The absolute trace of t is 0 in ZMod 2.

    Since the sum is 0, the algebra map sends the trace of t to 0 in E. Because ZMod 2 is a field, the algebra map into the nontrivial field E is injective, so the trace of t itself must be 0.

    lines 29–31

    uses FaithfulSMul.algebraMap_injective

      rw [hsum] at htrace
      have hinj := FaithfulSMul.algebraMap_injective (ZMod 2) E
      exact hinj (by simp [htrace])
FiniteField.algebraMap_trace_eq_sum_pow
The image of an element's trace under the canonical embedding of a finite base field K into a finite extension field L is the sum of the element's Frobenius conjugates, ∑ i < [L : K], x^(|K|^i).
FaithfulSMul.algebraMap_injective
The canonical homomorphism from a base ring to an algebra is injective when the scalar multiplication is faithful (as is always the case over a field).
Finset.sum_range_add
A sum over the range n + m can be split into the sum over the range n plus the sum over the range m shifted by n.
Finset.sum_congr
Two sums over the same finite set are equal if their terms agree pointwise.
Module.card_eq_pow_finrank
The cardinality of any finite-dimensional vector space V over a finite division ring K is |K|^(dim_K V).
Nat.pow_right_injective
For any base a ≥ 2, the function x ↦ a^x is injective on natural numbers.
ZMod.card
The cardinality of the ring ZMod n is n.
map_ofNat
A ring homomorphism maps the natural number literal n to the corresponding natural number literal n.
map_zero
Any zero-preserving map sends 0 to 0.
pow_add
In any monoid, a^(m + n) = a^m * a^n.
pow_mul
In any monoid, a^(m * n) = (a^m)^n.
two_mul
In any semiring, 2 * n = n + n.
zero_mul
Multiplying zero by any element gives zero.

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

/-- The absolute trace from an even-degree extension E of F_{2^m} vanishes on the subfield F_{2^m}. -/theorem subfield_element_trace_zero {E : Type*} [Field E] [Fintype E] [Algebra (ZMod 2) E] (m : ℕ) (hcard : Fintype.card E = 2 ^ (2 * m)) (t : E) (ht : t ^ (2 ^ m) = t) : Algebra.trace (ZMod 2) E t = 0 := by  have hrank : Module.finrank (ZMod 2) E = 2 * m := by    have h := Module.card_eq_pow_finrank (K := ZMod 2) (V := E)    rw [ZMod.card, hcard] at h    exact Nat.pow_right_injective (by norm_num) h.symm  have htrace := FiniteField.algebraMap_trace_eq_sum_pow (ZMod 2) E t  have hcard2 : Nat.card (ZMod 2) = 2 := by simp  rw [hrank, hcard2] at htrace  have hsum : ∑ i ∈ Finset.range (2 * m), t ^ (2 ^ i) = 0 := by    have h2m : 2 * m = m + m := two_mul m    rw [h2m, Finset.sum_range_add]    have h_eq : ∑ x ∈ Finset.range m, t ^ (2 ^ (m + x)) = ∑ x ∈ Finset.range m, t ^ (2 ^ x) := by      apply Finset.sum_congr rfl      intro x _      rw [pow_add, pow_mul, ht]    rw [h_eq]    have h2 : (2 : E) = 0 := by      have h : (2 : E) = algebraMap (ZMod 2) E (2 : ZMod 2) := (map_ofNat (algebraMap (ZMod 2) E) 2).symm      rw [h]      have : (2 : ZMod 2) = 0 := rfl      rw [this, map_zero]    have h_add : ∀ s : E, s + s = 0 := by      intro s      calc s + s = 2 * s := (two_mul s).symm      _ = 0 * s := by rw [h2]      _ = 0 := zero_mul s    exact h_add _  rw [hsum] at htrace  have hinj := FaithfulSMul.algebraMap_injective (ZMod 2) E  exact hinj (by simp [htrace])
See it in the graphDiscuss this resultChallenge it