The norm of a translation difference of a function is bounded by
ProvedCauchyRiemannIndex.eLpNorm_translate_sub_lemeasure-theoryreal-analysissobolev-spaces
Let , let be a real Banach space, let be continuously differentiable, and let . Then
where is the operator norm of the derivative of at .
This is the standard modulus-of-continuity bound for translations in in terms of the gradient. It gives equicontinuity of translations, uniformly over families with bounded norm, and is used for compactness statements for Sobolev-bounded families. No integrability of or is assumed; the right-hand side may be infinite.
Formalization Note. is in ℝ≥0∞; norms are written with eLpNorm for Lebesgue measure on EuclideanSpace ℝ (Fin m), is ENNReal.ofReal ‖h‖, is fderiv ℝ g and is ContDiff ℝ 1.
Preamble
import Mathlib.Analysis.Calculus.ContDiff.Basic import Mathlib.MeasureTheory.Function.LpSeminorm.Basic import Mathlib.MeasureTheory.Measure.Haar.InnerProductSpace
Formal statement
namespace CauchyRiemannIndex
open MeasureTheory
open scoped ENNReal NNReal
/-- For `1 ≤ p < ∞` and a `C¹` map `g : ℝᵐ → F` into a Banach space, the `Lᵖ` norm of the
translation difference `x ↦ g (x + h) - g x` is at most `|h|` times the `Lᵖ` norm of `|Dg|`. -/
theorem eLpNorm_translate_sub_le {m : ℕ} {F : Type*} [NormedAddCommGroup F] [NormedSpace ℝ F]
[CompleteSpace F] (p : ℝ≥0∞) (hp : 1 ≤ p) (hp' : p ≠ ⊤)
(g : EuclideanSpace ℝ (Fin m) → F) (hg : ContDiff ℝ 1 g) (h : EuclideanSpace ℝ (Fin m)) :
eLpNorm (fun x => g (x + h) - g x) p volume ≤
ENNReal.ofReal ‖h‖ * eLpNorm (fun x => ‖fderiv ℝ g x‖) p volume := by
sorry
end CauchyRiemannIndex
Source
C. Wendl, Lectures on Symplectic Field Theory, arXiv:1612.01009, https://arxiv.org/abs/1612.01009, Lecture 4, p. 74 (the compact inclusion used after Lemma 4.13); standard Sobolev space theory, e.g. H. Brezis, Functional Analysis, Sobolev Spaces and Partial Differential Equations, Springer 2011, Proposition 9.3 and Theorem 9.16