Directional derivatives commute with mollification of functions on
ProvedJHolomorphic.fderiv_normed_convolution_applyconvolutionmollifiersweak-derivatives
Let be with compact support, let be a bump function centred at and let be its normalization. Then for all , .
Formalization Note For a bump function centred at (ContDiffBump (0 : ℂ)), is ρ.normed volume, the multiple of with integral ; is the convolution ρ.normed volume ⋆[lsmul ℝ ℝ, volume] g with respect to Lebesgue measure on . The derivatives are fderiv ℝ.
Preamble
import Definitions.Def_JHolomorphic_Setting import Mathlib.Analysis.Calculus.ContDiff.Convolution import Mathlib.Analysis.Calculus.BumpFunction.Convolution import Mathlib.Analysis.Calculus.BumpFunction.FiniteDimension
Formal statement
namespace JHolomorphic
open MeasureTheory Filter Topology Metric ContinuousLinearMap
open scoped ContDiff Convolution
/-- Directional derivatives of a mollified `C¹` function are mollified directional derivatives. -/
theorem fderiv_normed_convolution_apply (ρ : ContDiffBump (0 : ℂ)) (φ : ℂ → ℝ)
(hφ : ContDiff ℝ 1 φ) (hc : HasCompactSupport φ) (z v : ℂ) :
fderiv ℝ (ρ.normed volume ⋆[lsmul ℝ ℝ, volume] φ) z v =
(ρ.normed volume ⋆[lsmul ℝ ℝ, volume] fun x => fderiv ℝ φ x v) z := by
sorry
end JHolomorphic
Source
Standard (mollification).