Literature Poincaré Bound at a Log-Concave Weak Limit
ProvedFunctionalInequalities.KLS.LogConcave.localSlope_constant_le_of_tendsto_readableThis formalizes the weak-limit transfer step of the regular-approximation reduction associated with Bizeul–Klartag–Lehec (arXiv:2610.05474v1), Section 2, (12), p. 5 (unnumbered).
Let be a finite-dimensional complete real inner product space with its Borel measurable structure and a second-countable topology. Assume:
- Probability measures converge weakly to a log-concave probability measure (
KLS.LogConcave). - and .
- Each satisfies
Poincare.SmoothTest.Holdswith real constant : every admissible smooth test satisfies . Admissibility requires and integrable squared gradient.
Then the intrinsic literature local-slope constant satisfies
This transfers supplied smooth-test bounds to the literature constant of the limit; the approximating laws need not be log-concave.
Formalization Note is Poincare.LocalSlope.Intrinsic.constant, valued in ENNReal. The constants are NNReal, coerced to reals in the hypotheses and to ENNReal in the conclusion.
Original Lean declaration header
Within namespace FunctionalInequalities.KLS.LogConcave. Shown in its original module context. Proof omitted.
theorem localSlope_constant_le_of_tendsto (μs : ℕ → ProbabilityMeasure E)
(μ : ProbabilityMeasure E) (hμ : LogConcave μ) (Cseq : ℕ → NNReal) (C : NNReal)
(hlim : Tendsto μs atTop (𝓝 μ)) (hClim : Tendsto Cseq atTop (𝓝 C))
(hbound : ∀ j, Poincare.SmoothTest.Holds (μs j : Measure E) (Cseq j : ℝ)) :
Poincare.LocalSlope.Intrinsic.constant μ ≤ (C : ENNReal)
Module navigation
Module: KLS/LogConcave/WeakLimit.
Definition modules: Probability/LogConcave, Poincare/LocalSlope/Basic, Poincare/SmoothTest/Basic.
Selected supporting theorems: exists_ambient_extension, Admissible.of_contDiff_hasCompactSupport, Holds.of_compactSupport, holds_iff_localSlope, variance_le_mul_energy_of_tendsto.
Source excerpt, lines 28–32 · Frozen source 408915f.
import Definitions.Def_KLSModular_FunctionalInequalities_KLS_LogConcave_WeakLimit
open MeasureTheory Filter Topology
variable {E : Type*} [NormedAddCommGroup E] [InnerProductSpace ℝ E]
[CompleteSpace E] [FiniteDimensional ℝ E] [MeasurableSpace E] [BorelSpace E]
[SecondCountableTopology E]
theorem FunctionalInequalities.KLS.LogConcave.localSlope_constant_le_of_tendsto_readable (μs : ℕ → ProbabilityMeasure E)
(μ : ProbabilityMeasure E) (hμ : LogConcave μ) (Cseq : ℕ → NNReal) (C : NNReal)
(hlim : Tendsto μs atTop (𝓝 μ)) (hClim : Tendsto Cseq atTop (𝓝 C))
(hbound : ∀ j, Poincare.SmoothTest.Holds (μs j : Measure E) (Cseq j : ℝ)) :
Poincare.LocalSlope.Intrinsic.constant μ ≤ (C : ENNReal) := by sorry