Borell Density Under a Concave Tilt
ProvedFunctionalInequalities.KLS.LogConcave.isBorellDensity_of_concave_tilt_modularThis states stability of Borell densities under continuous concave tilting.
Let be a real inner-product space with Borel measurable structure. Let and be probability measures, let be a measure, and let be a Borell density of with respect to . Let be continuous and concave on all of , and assume
Then has Borell density
with respect to . The definition permits zeros of . This result is a measure-theoretic closure property for the log-concave class and does not assume finite dimensionality in its Lean type.
Superseded upload. This snapshot was replaced by FunctionalInequalities.KLS.LogConcave.isBorellDensity_of_concave_tilt_readable: open the readable replacement. The replacement preserves the frozen source and mathematical statement, with a compact declaration context. This historical node remains available for existing imports.
import Definitions.Def_KLSModular_FunctionalInequalities_KLS_LogConcave_Density_ConcaveTilt
noncomputable section
open MeasureTheory Set Filter Topology FunctionalInequalities
section
namespace FunctionalInequalities
end FunctionalInequalities
open FunctionalInequalities
section
namespace FunctionalInequalities.KLS
end FunctionalInequalities.KLS
open FunctionalInequalities FunctionalInequalities.KLS
section
namespace FunctionalInequalities.KLS.LogConcave
end FunctionalInequalities.KLS.LogConcave
open FunctionalInequalities FunctionalInequalities.KLS FunctionalInequalities.KLS.LogConcave
variable {E : Type*} [NormedAddCommGroup E] [InnerProductSpace ℝ E]
[FiniteDimensional ℝ E] [MeasurableSpace E] [BorelSpace E]
omit [FiniteDimensional ℝ E]
theorem FunctionalInequalities.KLS.LogConcave.isBorellDensity_of_concave_tilt_modular (μ ν : ProbabilityMeasure E)
(m : Measure E) {ρ : E → ℝ} (hρ : KLS.IsBorellDensity (μ : Measure E) m ρ)
(φ : E → ℝ) (hφ : Continuous φ) (hconc : ConcaveOn ℝ univ φ)
(hν : (ν : Measure E) = (μ : Measure E).tilted φ) :
KLS.IsBorellDensity (ν : Measure E) m
(fun x => ρ x * (Real.exp (φ x) / ∫ y, Real.exp (φ y) ∂(μ : Measure E))) := by sorry