Proposition 2.13 -- irreducibility of a group walk
OpenMarkovMixing.group_walk_irreducible_iffThe random walk on a finite group with increment distribution is irreducible if and only if the support generates .
import Definitions.Def_mm_basic
namespace MarkovMixing
/-- **Proposition 2.13** (LPW): the random walk on a finite group with
increment distribution `μ` is irreducible if and only if the support of `μ`
generates the group. -/
theorem group_walk_irreducible_iff {G : Type*} [Group G] [Fintype G]
[DecidableEq G] (μ : G → ℝ) (hμ : IsDist μ) :
Irreducible (groupWalk μ) ↔ Subgroup.closure {g : G | 0 < μ g} = ⊤ := by
sorry
end MarkovMixing
Read-back
What the Lean code literally says, in plain math · claude-fable-5
Let be a finite group (with decidable equality on its elements), and let be a function satisfying the bundle's IsDist predicate, i.e. for every and . Define the real matrix by the bundle's definition
(note the order: the entry from to is evaluated at ). The theorem asserts the following equivalence: satisfies the bundle's Irreducible predicate — meaning that for every pair of elements there exists a natural number such that the entry of the matrix power is strictly positive, (the exponent is allowed to be , in which case is the identity matrix, so for the condition is automatically satisfied by ) — if and only if the subgroup of generated by the support set is the whole group (i.e. the subgroup closure of the set of elements where is strictly positive equals the top subgroup ). No aperiodicity, symmetry, or non-degeneracy condition on beyond nonnegativity and summing to is assumed; the statement is an if-and-only-if, quantified over the fixed but arbitrary finite group and distribution .
Confirmed by the mission captain (proposal self-audit).