The aggregation depth of an odd modulus is its Carmichael exponent.
ProvedRound10.least_complete_exponent_oddaether-cataloggeometry
The aggregation depth of an odd modulus is its Carmichael exponent. The least
positive exponent whose free witness is complete is λ(N) = lcm_{p ∣ N} φ(p^{v_p(N)}).
theorem Round10.least_complete_exponent_odd{N : ℕ} (hodd : Odd N) (hN : N ≠ 0) :
IsLeast {m : ℕ | 0 < m ∧ freeWitness N m = Nat.totient N}
(N.primeFactors.lcm fun p => Nat.totient (p ^ N.factorization p)) := by sorry
Formalization Note Transplanted verbatim from the Aether Catalog source Geometry/Round10Closures/CarmichaelThreshold.lean; the statement is byte-identical to the source declaration, elaborated with autoImplicit disabled in the platform environment.
Preamble
-- Thm stub generated from Geometry/Round10Closures/CarmichaelThreshold.lean
import Mathlib
import Definitions.Def_Geometry_Round10Closures_TraceLemma
/-
Round-10 Closures — Part IX (cycle 5): the aggregation depth of an arbitrary odd modulus.
Cycle 4 identified the exact aggregation depth of the free-witness channel for semiprimes
and for squarefree moduli: the least exponent with a maximal witness is `lcm_{r ∣ N}(r-1)`.
Cycle 5 removes the squarefree hypothesis on the odd part: for *every* odd `N`,
R_k(N) = ∏_{p ∣ N} gcd(φ(p^{v_p(N)}), k),
and the least positive exponent with `R_k(N) = φ(N)` is the Carmichael exponent
`λ(N) = lcm_{p ∣ N} φ(p^{v_p(N)})`.
The only input beyond the previous cycles is the cyclicity of `(ZMod (p^e))ˣ` for odd
primes; the `2`-adic case is genuinely different (the local group is not cyclic for
`8 ∣ N`) and is left open.
-/
open Round10Formal statement
theorem Round10.least_complete_exponent_odd{N : ℕ} (hodd : Odd N) (hN : N ≠ 0) :
IsLeast {m : ℕ | 0 < m ∧ freeWitness N m = Nat.totient N}
(N.primeFactors.lcm fun p => Nat.totient (p ^ N.factorization p)) := by sorrySource