AFTD Auto-Formalizing Theoretical Domains

slow_mode_dephasing

def

For permutations in the symmetric group S_n, with H the average of left-regular permutation matrices for all m-cycles, kappa_{n,m}(s) is the sum over g in S_n of |(exp(-i * s * H))_{g, 1}|^2 * (fix(g) - 1) / (n - 1), where fix(g) is the number of fixed points of g in {0, ..., n-1}.

Statement

noncomputable def slow_mode_dephasing (n m : ℕ) (s : ℝ) : ℝ
Defined in
arXiv:2609.13387, Eq. (89)
Verified
11 Nov 2026
Axioms
none listed
Used by 1
  • slow_mode_dephasing_zeroAt time s = 0, the quantum slow-mode dephasing factor kappa_{n,m}(0) equals 1 for every n >= 2 and every m.

Read back from the Lean

For natural numbers n and m, and a real number s, slow_mode_dephasing n m s is the real number defined as follows: 1. Let H be the (Equiv.Perm (Fin n)) × (Equiv.Perm (Fin n)) complex matrix whose (g, h)-entry is: - (|C|)⁻¹ (as a complex number) if (g * h⁻¹).cycleType = {m}, where C is the finite set of all permutations c : Equiv.Perm (Fin n) whose cycleType is the singleton multiset {m}; - 0 otherwise. 2. Compute the matrix exponential exp(- (I * s) • H). 3. Sum over all permutations g : Equiv.Perm (Fin n) of the product of: - the squared complex norm ‖(exp(- (I * s) • H)) g 1‖^2 of the entry at row g and column 1 (the identity permutation); - the normalized character (fix(g) - 1) / (n - 1), where fix(g) is the number of fixed points #{i : Fin n | g i = i}, cast to ℝ.

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

Lean source view module on GitHub

/-- Standard-irrep dephasing factor kappa_{n,m}(s) = sum_{g in S_n} |(exp(-i s H))_{g,1}|^2 (fix(g) - 1)/(n - 1) for the continuous-time quantum walk on S_n generated by m-cycles. -/
noncomputable def slow_mode_dephasing (n m : ℕ) (s : ℝ) : ℝ :=
  let H : Matrix (Equiv.Perm (Fin n)) (Equiv.Perm (Fin n)) ℂ :=
    fun g h ↦ if (g * h⁻¹).cycleType = {m} then
      ((({c : Equiv.Perm (Fin n) | c.cycleType = {m}} : Finset (Equiv.Perm (Fin n))).card : ℂ)⁻¹)
    else 0
  ∑ g : Equiv.Perm (Fin n),
    ‖(NormedSpace.exp (- (Complex.I * (s : ℂ)) • H)) g 1‖ ^ 2 *
      ((((Finset.filter (fun i : Fin n ↦ g i = i) Finset.univ).card : ℝ) - 1) / ((n : ℝ) - 1))
See it in the graphDiscuss this resultChallenge it