AFTD Auto-Formalizing Theoretical Domains

cycle_power_dyadic_only_coloured_cover_counterexample

theoremverified

At radius eight and period twenty, four real class colours can coexist with tight classes, constant active count four, exact coverage of all dyadic long edges and disjoint classes of each colour, yet an ordinary distance-one edge is uncovered. Thus keeping the palette does not make dyadic-only coverage force divisibility.

Statement

theorem cycle_power_dyadic_only_coloured_cover_counterexample :
    ∃ (x : ZMod 20 → ℕ → Bool) (ell : ZMod 20 → ℕ),
      (∀ p, ell p < 4) ∧
      (∀ p, x p 0 = true ∧ x p 7 = true) ∧
      (∀ p, cycle_power_dyadic_class_weight 3 (x p) =
        (Finset.range 8).sum (fun a => (x p a).toNat)) ∧
      (∀ v : ZMod 20, (Finset.range 8).sum (fun a =>
        (x (v - (a : ZMod 20)) a).toNat) = 4) ∧
      (∀ (j : Fin 3) (v : ZMod 20),
        (Finset.range (2 ^ (j : ℕ))).sum (fun a =>
          (x (v - (a : ZMod 20)) a &&
            x (v - (a : ZMod 20)) (a + 8 - 2 ^ (j : ℕ))).toNat) = 1) ∧
      (∀ (v : ZMod 20) (a b : Fin 8), a ≠ b →
        x (v - ((a : ℕ) : ZMod 20)) a = true →
        x (v - ((b : ℕ) : ZMod 20)) b = true →
        ell (v - ((a : ℕ) : ZMod 20)) ≠ ell (v - ((b : ℕ) : ZMod 20))) ∧
      (∀ a ∈ Finset.range 7,
        ¬(x ((2 : ZMod 20) - (a : ZMod 20)) a = true ∧
          x ((2 : ZMod 20) - (a : ZMod 20)) (a + 1) = true))
Source
Counterexample to a proposed shortcut in OP-205 / RT-1, following arXiv:2610.15635. The missing ordinary edge is explicit, so this is not a counterexample to the paper or the main full-cover conjecture. Experiment6382; no research advance claimed.
Verified
27 Oct 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 1
  • cycle_power_dyadic_class_weightThe weighted number of pairs in a dyadic arc: the endpoint pair has weight two, and distances 2^m-2^j for 1≤j<m have weight one.

The proof, explained

The proof is by direct construction of an explicit counterexample. An indicator configuration $x : \mathbb{Z}/20\mathbb{Z} \to \mathbb{N} \to \text{Bool}$ and a 4-colouring $\ell : \mathbb{Z}/20\mathbb{Z} \to \{0, 1, 2, 3\}$ are defined explicitly by lookup tables for each class modulo 20. Once $x$ and $\ell$ are provided as witnesses, the seven required conditions—palette size at most 4, boundary tightness, dyadic class weight matching, constant active degree 4, exact dyadic coverage, disjointness of classes sharing a colour, and an uncovered distance-one edge at vertex 2—reduce to a decidable proposition over finite domains and are verified by exhaustive finite computation (`decide`).

  1. Define the configuration function $x : \mathbb{Z}/20\mathbb{Z} \to \mathbb{N} \to \text{Bool}$

    For each class $p \in \mathbb{Z}/20\mathbb{Z}$, the active elements in $\{0, \dots, 7\}$ are explicitly specified via a pattern match on $p$.

    lines 23–44

      let x : ZMod 20 → ℕ → Bool := fun p a => decide (a ∈
        match p.val with
        | 0 => ({0, 1, 3, 4, 6, 7} : Finset ℕ)
        | 1 => {0, 1, 5, 7}
        | 2 => {0, 3, 7}
        | 3 => {0, 1, 6, 7}
        | 4 => {0, 4, 7}
        | 5 => {0, 2, 6, 7}
        | 6 => {0, 1, 3, 4, 6, 7}
        | 7 => {0, 1, 5, 7}
        | 8 => {0, 3, 7}
        | 9 => {0, 1, 5, 6, 7}
        | 10 => {0, 7}
        | 11 => {0, 2, 6, 7}
        | 12 => {0, 1, 2, 4, 6, 7}
        | 13 => {0, 7}
        | 14 => {0, 1, 3, 5, 6, 7}
        | 15 => {0, 1, 5, 7}
        | 16 => {0, 7}
        | 17 => {0, 1, 2, 6, 7}
        | 18 => {0, 4, 7}
        | _ => {0, 2, 6, 7})
  2. Define the class colouring $\ell : \mathbb{Z}/20\mathbb{Z} \to \mathbb{N}$

    Each class $p \in \mathbb{Z}/20\mathbb{Z}$ is assigned an explicit colour index in $\{0, 1, 2, 3\}$ via a pattern match on $p$.

    lines 45–66

      let ell : ZMod 20 → ℕ := fun p =>
        match p.val with
        | 0 => 0
        | 1 => 2
        | 2 => 0
        | 3 => 1
        | 4 => 3
        | 5 => 2
        | 6 => 3
        | 7 => 1
        | 8 => 0
        | 9 => 2
        | 10 => 0
        | 11 => 1
        | 12 => 0
        | 13 => 2
        | 14 => 3
        | 15 => 1
        | 16 => 3
        | 17 => 2
        | 18 => 3
        | _ => 1
  3. It suffices to verify that the explicit witnesses $x$ and $\ell$ satisfy the seven required properties

    Instantiate the existential quantifier $\exists x\ \ell, \dots$ with the candidates defined above.

    line 67

      refine ⟨x, ell, ?_⟩
  4. The candidate pair $(x, \ell)$ satisfies all seven conditions, completing the proof automation

    Because the variables range over finite types ($\mathbb{Z}/20\mathbb{Z}$, $\text{Fin } 3$, $\text{Fin } 8$, and finite ranges) and all relations are computable, the entire conjunction is a decidable proposition verified by exhaustive finite computation.

    line 68

      decide

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

set_option maxRecDepth 10000set_option maxHeartbeats 8000000theorem cycle_power_dyadic_only_coloured_cover_counterexample :    ∃ (x : ZMod 20 → ℕ → Bool) (ell : ZMod 20 → ℕ),      (∀ p, ell p < 4) ∧      (∀ p, x p 0 = true ∧ x p 7 = true) ∧      (∀ p, cycle_power_dyadic_class_weight 3 (x p) =        (Finset.range 8).sum (fun a => (x p a).toNat)) ∧      (∀ v : ZMod 20, (Finset.range 8).sum (fun a =>        (x (v - (a : ZMod 20)) a).toNat) = 4) ∧      (∀ (j : Fin 3) (v : ZMod 20),        (Finset.range (2 ^ (j : ℕ))).sum (fun a =>          (x (v - (a : ZMod 20)) a &&            x (v - (a : ZMod 20)) (a + 8 - 2 ^ (j : ℕ))).toNat) = 1) ∧      (∀ (v : ZMod 20) (a b : Fin 8), a ≠ b →        x (v - ((a : ℕ) : ZMod 20)) a = true →        x (v - ((b : ℕ) : ZMod 20)) b = true →        ell (v - ((a : ℕ) : ZMod 20)) ≠ ell (v - ((b : ℕ) : ZMod 20))) ∧      (∀ a ∈ Finset.range 7,        ¬(x ((2 : ZMod 20) - (a : ZMod 20)) a = true ∧          x ((2 : ZMod 20) - (a : ZMod 20)) (a + 1) = true)) := by  let x : ZMod 20 → ℕ → Bool := fun p a => decide (a ∈    match p.val with    | 0 => ({0, 1, 3, 4, 6, 7} : Finset ℕ)    | 1 => {0, 1, 5, 7}    | 2 => {0, 3, 7}    | 3 => {0, 1, 6, 7}    | 4 => {0, 4, 7}    | 5 => {0, 2, 6, 7}    | 6 => {0, 1, 3, 4, 6, 7}    | 7 => {0, 1, 5, 7}    | 8 => {0, 3, 7}    | 9 => {0, 1, 5, 6, 7}    | 10 => {0, 7}    | 11 => {0, 2, 6, 7}    | 12 => {0, 1, 2, 4, 6, 7}    | 13 => {0, 7}    | 14 => {0, 1, 3, 5, 6, 7}    | 15 => {0, 1, 5, 7}    | 16 => {0, 7}    | 17 => {0, 1, 2, 6, 7}    | 18 => {0, 4, 7}    | _ => {0, 2, 6, 7})  let ell : ZMod 20 → ℕ := fun p =>    match p.val with    | 0 => 0    | 1 => 2    | 2 => 0    | 3 => 1    | 4 => 3    | 5 => 2    | 6 => 3    | 7 => 1    | 8 => 0    | 9 => 2    | 10 => 0    | 11 => 1    | 12 => 0    | 13 => 2    | 14 => 3    | 15 => 1    | 16 => 3    | 17 => 2    | 18 => 3    | _ => 1  refine ⟨x, ell, ?_⟩  decide
See it in the graphDiscuss this resultChallenge it