Cross entropy bounds entropy for a subprobability comparison
Provedmme_entropy_le_cross_entropymatrix-multiplicationregional-extraction
For finite probability distributions, a nonnegative comparison function of total mass at most one bounds entropy by cross entropy whenever it is positive on the probability support. The full matrix exponent bound remains a separate obligation.
Preamble
import Definitions.Def_mme_regional_entropy_rate_data import Mathlib.Analysis.SpecialFunctions.Log.Basic import Mathlib.Tactic.FieldSimp import Mathlib.Tactic.Linarith open scoped BigOperators open MME.RegionRate
Formal statement
theorem mme_entropy_le_cross_entropy {W : Type*} [Fintype W]
(p q : W → ℝ) (hp : ∀ w, 0 ≤ p w) (hq : ∀ w, 0 ≤ q w)
(hmass : ∑ w, p w = 1) (hqmass : ∑ w, q w ≤ 1)
(hsupport : ∀ w, 0 < p w → 0 < q w) :
entropy p ≤ -∑ w, p w * Real.log (q w) := by sorrySource
Kernel-checked regional count identities and exact extraction with empty regions.