Theorem 1.8 — Kolmogorov's criterion: a stationary Markov process is reversible iff (1.22) holds
ProvedKellyReversibility.Reversibility.process_kolmogorov_criterionLet , , be a stationary Markov process on a finite state space with transition rates (), irreducible, with equilibrium distribution . The process is reversible if and only if its transition rates satisfy
for any finite sequence of states .
Reversibility, a property of the joint laws of the process at all finite sets of times, is thus decided by the rates alone: a reversible process shows no net circulation around any cycle of states.
Formalization Note Reversibility is the distributional definition of p. 5: the finite-dimensional distributions are invariant under . It is not defined as detailed balance, and two-way communication () is not assumed. The sequences in (1.22) have arbitrary length and may repeat states. The state space is finite (Kelly allows a countable one).
Retired 2026-10-27: a Lean-vs-paper fidelity audit found this statement weaker than, or narrower than, the cited result. A corrected statement will be resubmitted.
import Mathlib import Definitions.Def_KellyStochasticNetworks_Balance import Definitions.Def_KellyReversibility_Reversibility_StationaryLaw import Definitions.Def_KellyReversibility_Reversibility_MarkovProcess
namespace KellyReversibility.Reversibility
/-- Theorem 1.8 (Kelly, p. 23). -/
theorem process_kolmogorov_criterion {S : Type*} [Fintype S] [DecidableEq S]
(q : S → S → ℝ) (hq : ∀ j k, j ≠ k → 0 ≤ q j k) (hq0 : ∀ j, q j j = 0)
(hirr : RatesIrreducible q)
(π₀ : S → ℝ) (hπ₀ : IsEquilibrium π₀ q) :
ProcessReversible q π₀ ↔ KolmogorovCycle q := by sorry
end KellyReversibility.Reversibility
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.