Nonnegative entropy production for finite stationary Markov chains
ProvedClassicalGaps.entropy_production_nonnegentropy-productionmarkov-chainssecond-lawstochastic-systems
Let be a finite CTMC generator with strictly positive off-diagonal rates and let be a strictly positive stationary distribution. Then the total entropy production, the sum over ordered pairs of the stationary current imbalance times the log ratio of forward to reverse currents, is nonnegative.
Preamble
import Mathlib.Analysis.SpecialFunctions.Log.Basic import Mathlib.Data.Finset.Erase import Mathlib.Algebra.BigOperators.Group.Finset.Defs open scoped BigOperators
Formal statement
theorem ClassicalGaps.entropy_production_nonneg {V : Type*} [Fintype V] [DecidableEq V]
(π : V → ℝ) (Q : Matrix V V ℝ)
(hoff : ∀ i j, i ≠ j → 0 < Q i j) (hrowsum : ∀ i, ∑ j, Q i j = 0)
(hstat : ∀ j, ∑ i, π i * Q i j = 0) (hπ : ∀ i, 0 < π i) :
0 ≤ ∑ i, ∑ j ∈ Finset.univ.erase i,
(π i * Q i j - π j * Q j i) * Real.log (π i * Q i j / (π j * Q j i)) := by sorrySource
J. Schnakenberg, Network theory of microscopic and macroscopic behavior of master equation systems, Rev. Mod. Phys. 48 (1976) 571; see https://en.wikipedia.org/wiki/Entropy_production