Two-copy entropy inequality for a finite linear supremum process
ProvedTalagrandCore.two_copy_entropy_Zconcentration-inequalitiesempirical-processesprobabilitytalagrand
For a centered finite Bernoulli linear supremum with coefficient envelope one and variance proxy , define the random variance process . For every ,
This is the two-copy entropy estimate feeding the Gaussian upper-tail differential inequality.
Formalization Note Suprema are finite maxima, so measurable-selection issues are replaced by a finite argmax construction inside the proof.
Preamble
import Definitions.Def_talagrand_finite_bool_core open MeasureTheory open scoped Classical BigOperators
Formal statement
namespace TalagrandCore
variable {κ ι : Type} [DecidableEq κ] [Fintype κ]
variable [Fintype ι] [Nonempty ι]
theorem two_copy_entropy_Z (p : NNReal) (hp : p ≤ 1)
(coeff : ι → κ → ℝ) (hB : ∀ a x, |coeff a x| ≤ 1)
(sigmaSq : ℝ)
(hVar : ∀ a : ι, ∑ x : κ, (p : ℝ) * (1 - (p : ℝ)) * coeff a x ^ 2 ≤ sigmaSq)
(lam : ℝ) (hlam : 0 ≤ lam) :
lam * Ex (p : ℝ) (fun ω => Zproc coeff (p : ℝ) ω *
Real.exp (lam * Zproc coeff (p : ℝ) ω)) -
Ex (p : ℝ) (fun ω => Real.exp (lam * Zproc coeff (p : ℝ) ω)) *
Real.log (Ex (p : ℝ) (fun ω => Real.exp (lam * Zproc coeff (p : ℝ) ω))) ≤
lam ^ 2 * Real.exp (2 * lam) *
Ex (p : ℝ) (fun ω => Real.exp (lam * Zproc coeff (p : ℝ) ω) *
(Sigma2 coeff (p : ℝ) ω + sigmaSq)) := by sorry
end TalagrandCoreSource
Michel Ledoux, On Talagrand’s deviation inequalities for product measures, ESAIM Probability and Statistics 1 (1996), Theorems 2.4–2.5, pp. 63–87. Formal Lean proof extracted from Prove2Me accepted submission bf106d23-ff42-48f5-a837-a8528101c849 by tianyipeng.