AFTD Auto-Formalizing Theoretical Domains

cycle_power_dyadic_reflect_defective_half_model

theoremverified

Reflecting a defective half model reverses its defect function and changes its top-owner phase from shift to half-span minus shift.

Statement

theorem cycle_power_dyadic_reflect_defective_half_model (t : ℕ) (ht : 2 ≤ t) (y : ℤ → ℕ → Bool) (e : ℤ → Bool) (shift : ℤ) (hmodel : cycle_power_dyadic_defective_half_model t y e shift) : cycle_power_dyadic_defective_half_model t (cycle_power_rooted_reflect (2^t) y) (fun v => e (-v)) ((2 : ℤ)^(t-1) - shift)
Source
Reflection bridge between the two phase conventions in RT-1.
Verified
11 Nov 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 8
Used by 1

The proof, explained

The proof shows that the rooted reflection of a dyadic defective half model is another defective half model, with defect function $v \mapsto e(-v)$ and top-owner phase updated to $2^{t-1} - \text{shift}$. After decomposing the half-model definition into its five axioms, the class weight condition follows from sum reversal, lower-scale pair coverage and total vertex count transfer via general reflection lemmas, top-scale pair coverage transfers using the identity $2^t - 2^{t-1} = 2^{t-1}$, and chord coverage is verified by reflecting the chord witness coordinate $a$ to $2^t - d - 1 - a$.

  1. Unpack the defective half-model assumptions, establish dyadic arithmetic facts, and reduce to verifying the five model axioms for the reflected configuration.

    We decompose the hypothesis `hmodel` into its five properties: the dyadic class weight condition, exact pair coverage at lower scales $j < t - 1$, defective pair coverage at top scale $t - 1$, chord coverage, and the total vertex degree count. We note that $2^t > 0$ and $2^{t-1} + 2^{t-1} = 2^t$, then split the goal into the five corresponding obligations for the reflected model.

    lines 2–8

    uses Nat.two_pow_pos, pow_succ

      obtain ⟨hweight, hexact, htop, hcover, hcount⟩ := hmodel
      have hp : 0 < (2:ℕ)^t := Nat.two_pow_pos t
      have hhalf : 2^(t-1) + 2^(t-1) = (2:ℕ)^t := by
        have h : t-1+1=t := by omega
        conv_rhs => rw [← h, pow_succ]
        omega
      refine ⟨?_, ?_, ?_, ?_, ?_⟩
  2. For each phase $p$, the dyadic class weight of the reflected pattern equals its coordinate sum.

    By definition of rooted reflection, the reflected pattern at phase $p$ is obtained from $y$ at phase $-p - (2^t - 1)$ by reversing coordinate indices ($a \mapsto 2^t - 1 - a$). By `cycle_power_dyadic_class_weight_reflect`, reflection preserves the dyadic class weight. Applying the original weight hypothesis (`hweight`) and reversing the summation order over $[0, 2^t)$ via `Finset.sum_range_reflect`, the sum matches the coordinate sum of the reflected configuration.

    lines 9–16

    uses cycle_power_dyadic_class_weight_reflect, Finset.sum_range_reflect, Finset.sum_congr

      · intro p
        change cycle_power_dyadic_class_weight t (fun a => y (-p - ((2^t-1 : ℕ) : ℤ)) (2^t-1-a)) = _
        rw [cycle_power_dyadic_class_weight_reflect, hweight]
        rw [← Finset.sum_range_reflect]
        apply Finset.sum_congr rfl
        intro a ha
        have ha' : a < 2^t := Finset.mem_range.mp ha
        congr 2
  3. Exact pair coverage holds for all lower dyadic scales $j$ with $j + 1 < t$.

    For each scale $j$ with $j + 1 < t$, we apply the general lemma `cycle_power_rooted_reflect_exact` with span $D = 2^t$ and block length $L = 2^j$. The prerequisites $0 < 2^j$ and $2^j \le 2^t$ follow from `Nat.two_pow_pos` and monotonicity of powers (`Nat.pow_le_pow_right`), and `hexact` provides exact coverage in the original model.

    lines 17–19

    uses cycle_power_rooted_reflect_exact, Nat.two_pow_pos, Nat.pow_le_pow_right

      · intro j hj
        exact cycle_power_rooted_reflect_exact (2^t) (2^j) (Nat.two_pow_pos j)
          (Nat.pow_le_pow_right (by norm_num) (by omega)) y (hexact j hj)
  4. At the top scale $2^{t-1}$, defective pair coverage holds with phase parameter $2^{t-1} - \text{shift}$.

    Applying `cycle_power_reflect_shifted_pair_defect` with $D = 2^t$ and $L = 2^{t-1}$ shows that the reflected pair sum plus $e(-v - (2^t - 2^{t-1}) + \text{shift})$ equals 1. Because $2^t - 2^{t-1} = 2^{t-1}$, the defect term is $e(-v - 2^{t-1} + \text{shift})$. For the reflected defect function $v' \mapsto e(-v')$ shifted by $2^{t-1} - \text{shift}$, evaluating at $v + (2^{t-1} - \text{shift})$ gives $e(-(v + 2^{t-1} - \text{shift})) = e(-v - 2^{t-1} + \text{shift})$, which matches by ring arithmetic.

    lines 20–28

    uses cycle_power_reflect_shifted_pair_defect, Nat.two_pow_pos

      · intro v
        have h := cycle_power_reflect_shifted_pair_defect (2^t) (2^(t-1))
          (Nat.two_pow_pos _) (by omega) y e shift htop v
        have he : ((2^t - 2^(t-1) : ℕ) : ℤ) = (2 : ℤ)^(t-1) := by
          have hn : 2^t - 2^(t-1) = (2:ℕ)^(t-1) := by omega
          rw [hn]
          norm_cast
        rw [he] at h
        convert h using 1 <;> congr 3 <;> ring
  5. Every chord length $1 \le d < 2^t$ with $d \ne 2^{t-1}$ is covered by some position in the reflected model.

    Given chord length $d$ and vertex $v$, applying the chord coverage hypothesis `hcover` at vertex $-v - d$ provides a coordinate $a < 2^t - d$ where $y$ is true at both $a$ and $a + d$. Setting $b = 2^t - d - 1 - a$, we obtain $b < 2^t - d$. Under rooted reflection, $b$ maps to $2^t - 1 - b = a + d$ and $b + d$ maps to $2^t - 1 - (b + d) = a$, swapping the roles of the lower and higher endpoints. The vertex alignment $-(v - b) - (2^t - 1) = -v - d - a$ ensures both reflected endpoints are true.

    lines 29–39

      · intro d hd hdD hne v
        obtain ⟨a,ha,hlo,hhi⟩ := hcover d hd hdD hne (-v-(d : ℤ))
        let b := 2^t-d-1-a
        have hb : b < 2^t-d := by dsimp [b]; omega
        have hi₁ : 2^t-1-b=a+d := by dsimp [b]; omega
        have hi₂ : 2^t-1-(b+d)=a := by dsimp [b]; omega
        have hv : -(v-(b : ℤ))-((2^t-1 : ℕ) : ℤ) = -v-(d : ℤ)-(a : ℤ) := by
          dsimp [b]; omega
        refine ⟨b,hb,?_,?_⟩
        · simpa only [cycle_power_rooted_reflect,hi₁,hv] using hhi
        · simpa only [cycle_power_rooted_reflect,hi₂,hv] using hlo
  6. The total vertex degree count remains $t + 1$.

    Lemma `cycle_power_reflect_count_defect` directly establishes that rooted reflection preserves the total vertex sum plus the defect at $-v$, so the total count remains $t + 1$ by `hcount`.

    line 40

    uses cycle_power_reflect_count_defect

      · exact cycle_power_reflect_count_defect (2^t) hp y e (t+1) hcount
Nat.two_pow_pos
Powers of 2 are strictly positive: $2^w > 0$ for any natural number $w$.
pow_succ
In any monoid, $a^{n+1} = a^n \cdot a$.
cycle_power_dyadic_class_weight_reflect
Reversing a Boolean sequence of length $2^t$ preserves its dyadic class weight.
Finset.sum_range_reflect
Reversing the order of summation over the range $\{0, \dots, n-1\}$ leaves the sum unchanged.
Finset.sum_congr
Two finite sums are equal if their index sets are equal and their summands agree on every element.
Nat.pow_le_pow_right
For a positive base $n$, exponentiation is monotonic in the exponent: $i \le j$ implies $n^i \le n^j$.
cycle_power_rooted_reflect_exact
Rooted reflection preserves the exact pair coverage condition for a pair system of span $D$ with chord parameter $L$.
cycle_power_reflect_shifted_pair_defect
Rooted reflection preserves pair coverage with a shifted defect function, transforming the defect term from $v + \text{shift}$ to $-v - (D - L) + \text{shift}$.
cycle_power_reflect_count_defect
Rooted reflection preserves the total vertex degree count with a defect term, reversing the vertex argument to evaluate the defect at $-v$.

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

theorem cycle_power_dyadic_reflect_defective_half_model (t : ℕ) (ht : 2 ≤ t) (y : ℤ → ℕ → Bool) (e : ℤ → Bool) (shift : ℤ) (hmodel : cycle_power_dyadic_defective_half_model t y e shift) : cycle_power_dyadic_defective_half_model t (cycle_power_rooted_reflect (2^t) y) (fun v => e (-v)) ((2 : ℤ)^(t-1) - shift) := by  obtain ⟨hweight, hexact, htop, hcover, hcount⟩ := hmodel  have hp : 0 < (2:ℕ)^t := Nat.two_pow_pos t  have hhalf : 2^(t-1) + 2^(t-1) = (2:ℕ)^t := by    have h : t-1+1=t := by omega    conv_rhs => rw [← h, pow_succ]    omega  refine ⟨?_, ?_, ?_, ?_, ?_⟩  · intro p    change cycle_power_dyadic_class_weight t (fun a => y (-p - ((2^t-1 : ℕ) : ℤ)) (2^t-1-a)) = _    rw [cycle_power_dyadic_class_weight_reflect, hweight]    rw [← Finset.sum_range_reflect]    apply Finset.sum_congr rfl    intro a ha    have ha' : a < 2^t := Finset.mem_range.mp ha    congr 2  · intro j hj    exact cycle_power_rooted_reflect_exact (2^t) (2^j) (Nat.two_pow_pos j)      (Nat.pow_le_pow_right (by norm_num) (by omega)) y (hexact j hj)  · intro v    have h := cycle_power_reflect_shifted_pair_defect (2^t) (2^(t-1))      (Nat.two_pow_pos _) (by omega) y e shift htop v    have he : ((2^t - 2^(t-1) : ℕ) : ℤ) = (2 : ℤ)^(t-1) := by      have hn : 2^t - 2^(t-1) = (2:ℕ)^(t-1) := by omega      rw [hn]      norm_cast    rw [he] at h    convert h using 1 <;> congr 3 <;> ring  · intro d hd hdD hne v    obtain ⟨a,ha,hlo,hhi⟩ := hcover d hd hdD hne (-v-(d : ℤ))    let b := 2^t-d-1-a    have hb : b < 2^t-d := by dsimp [b]; omega    have hi₁ : 2^t-1-b=a+d := by dsimp [b]; omega    have hi₂ : 2^t-1-(b+d)=a := by dsimp [b]; omega    have hv : -(v-(b : ℤ))-((2^t-1 : ℕ) : ℤ) = -v-(d : ℤ)-(a : ℤ) := by      dsimp [b]; omega    refine ⟨b,hb,?_,?_⟩    · simpa only [cycle_power_rooted_reflect,hi₁,hv] using hhi    · simpa only [cycle_power_rooted_reflect,hi₂,hv] using hlo  · exact cycle_power_reflect_count_defect (2^t) hp y e (t+1) hcount
See it in the graphDiscuss this resultChallenge it