(2.2) — the canonical form of reversible transition rates
ProvedSerfozoStochasticNetworks.reversible_canonical_formLet be a non-negative rate function on a state space and let be strictly positive. Then satisfies the detailed balance equations for ,
if and only if there is a non-negative symmetric function with
So the rate functions reversible with respect to a given are exactly the symmetric kernels divided by . Besides being a test for reversibility, this is a recipe: any positive is the invariant measure of the reversible process built from any symmetric this way.
Formalization Note The book states the form for ; here it is asserted for all pairs, including the diagonal, which costs nothing — the diagonal value of is determined as and the detailed balance equation at is an identity.
is required to be non-negative, which is where non-negativity of is used; in the forward direction , and symmetry of that is precisely detailed balance.
The statement is about a fixed : it characterizes the rate functions reversible with respect to that , rather than asserting the existence of some reversible measure.
import Mathlib import Definitions.Def_SerfozoStochasticNetworks_Reversible
namespace SerfozoStochasticNetworks
theorem reversible_canonical_form {E : Type*} (q : E → E → ℝ) (π : E → ℝ)
(hq : ∀ x y, 0 ≤ q x y) (hπ : ∀ x, 0 < π x) :
DetailedBalance q π ↔ ∃ γ : E → E → ℝ, (∀ x y, 0 ≤ γ x y) ∧ (∀ x y, γ x y = γ y x) ∧
∀ x y, q x y = γ x y / π x := by sorry
end SerfozoStochasticNetworksConfirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.