Documentation

TauCeti.Analysis.Sobolev.W1p.Restriction

Restriction of first-order Sobolev functions #

A weakly differentiable function on an open set remains weakly differentiable on every smaller open set. This file packages that operation as the contractive continuous linear map TauCeti.W1p.restrictL. Its value and weak gradient are represented by the same functions on the smaller domain, and restriction is functorial.

Restriction is the basic localization operation for Sobolev spaces. In particular, it lets interior regularity arguments pass from a weak solution on Ω to smaller open sets.

Main declarations #

References #

L. C. Evans, Partial Differential Equations, Chapter 5, §5.2.

Restriction of ambient jets #

noncomputable def TauCeti.Sobolev1JetLp.restrictL {E : Type u_1} [MeasurableSpace E] [NormedAddCommGroup E] [InnerProductSpace ℝ E] {mu : MeasureTheory.Measure E} {Omega U : TopologicalSpace.Opens E} {p : ENNReal} [Fact (1 ≤ p)] (hU : U ≤ Omega) :
↥(Sobolev1JetLp mu Omega p) →L[ℝ] ↥(Sobolev1JetLp mu U p)

Restriction of an Lᵖ value-gradient jet to a smaller open set.

Equations
Instances For
    theorem TauCeti.Sobolev1JetLp.coeFn_restrictL {E : Type u_1} [MeasurableSpace E] [NormedAddCommGroup E] [InnerProductSpace ℝ E] {mu : MeasureTheory.Measure E} {Omega U : TopologicalSpace.Opens E} {p : ENNReal} [Fact (1 ≤ p)] (hU : U ≤ Omega) (J : ↥(Sobolev1JetLp mu Omega p)) :
    ↑↑((restrictL hU) J) =ᵐ[mu.restrict ↑U] ↑↑J

    Restricting an ambient jet keeps the same representative on the smaller open set.

    Restriction of ambient jets does not increase the Lᵖ norm.

    @[simp]

    The value component of a restricted ambient jet is the restriction of its value component.

    @[simp]

    The gradient component of a restricted ambient jet is the restriction of its gradient component.

    Restriction of Sobolev functions #

    noncomputable def TauCeti.W1p.restrictL {E : Type u_1} [MeasurableSpace E] [NormedAddCommGroup E] [InnerProductSpace ℝ E] [FiniteDimensional ℝ E] [BorelSpace E] {mu : MeasureTheory.Measure E} [mu.IsAddHaarMeasure] {Omega U : TopologicalSpace.Opens E} {p : ENNReal} [Fact (1 ≤ p)] (hU : U ≤ Omega) :
    ↥(W1p mu Omega p) →L[ℝ] ↥(W1p mu U p)

    Restriction of first-order Sobolev functions. If U ⊆ Ω, this is the continuous linear map W^{1,p}(Ω) → W^{1,p}(U) obtained by restricting both the value and weak gradient. It has operator norm at most one.

    Equations
    Instances For
      @[simp]
      theorem TauCeti.W1p.coe_restrictL {E : Type u_1} [MeasurableSpace E] [NormedAddCommGroup E] [InnerProductSpace ℝ E] [FiniteDimensional ℝ E] [BorelSpace E] {mu : MeasureTheory.Measure E} [mu.IsAddHaarMeasure] {Omega U : TopologicalSpace.Opens E} {p : ENNReal} [Fact (1 ≤ p)] (hU : U ≤ Omega) (u : ↥(W1p mu Omega p)) :
      ↑((restrictL hU) u) = (Sobolev1JetLp.restrictL hU) ↑u

      The ambient jet of a restricted Sobolev function is the restriction of its ambient jet.

      @[simp]

      The value component of a restricted Sobolev function is the restriction of its value component.

      @[simp]

      The gradient component of a restricted Sobolev function is the restriction of its gradient component.

      theorem TauCeti.W1p.value_restrictL_ae {E : Type u_1} [MeasurableSpace E] [NormedAddCommGroup E] [InnerProductSpace ℝ E] [FiniteDimensional ℝ E] [BorelSpace E] {mu : MeasureTheory.Measure E} [mu.IsAddHaarMeasure] {Omega U : TopologicalSpace.Opens E} {p : ENNReal} [Fact (1 ≤ p)] (hU : U ≤ Omega) (u : ↥(W1p mu Omega p)) :
      ↑↑(value ((restrictL hU) u)) =ᵐ[mu.restrict ↑U] ↑↑(value u)

      Restriction keeps the same value representative on the smaller open set.

      theorem TauCeti.W1p.gradient_restrictL_ae {E : Type u_1} [MeasurableSpace E] [NormedAddCommGroup E] [InnerProductSpace ℝ E] [FiniteDimensional ℝ E] [BorelSpace E] {mu : MeasureTheory.Measure E} [mu.IsAddHaarMeasure] {Omega U : TopologicalSpace.Opens E} {p : ENNReal} [Fact (1 ≤ p)] (hU : U ≤ Omega) (u : ↥(W1p mu Omega p)) :
      ↑↑(gradient ((restrictL hU) u)) =ᵐ[mu.restrict ↑U] ↑↑(gradient u)

      Restriction keeps the same weak-gradient representative on the smaller open set.

      Restriction does not increase the Sobolev norm.

      The restriction operator has norm at most one.

      @[simp]

      Restricting a Sobolev function to its original domain does nothing.

      @[simp]
      theorem TauCeti.W1p.restrictL_restrictL {E : Type u_1} [MeasurableSpace E] [NormedAddCommGroup E] [InnerProductSpace ℝ E] [FiniteDimensional ℝ E] [BorelSpace E] {mu : MeasureTheory.Measure E} [mu.IsAddHaarMeasure] {Omega U V : TopologicalSpace.Opens E} {p : ENNReal} [Fact (1 ≤ p)] (hU : U ≤ Omega) (hV : V ≤ U) (u : ↥(W1p mu Omega p)) :
      (restrictL hV) ((restrictL hU) u) = (restrictL ⋯) u

      Restriction is functorial: restricting from Ω to U and then to V agrees with direct restriction from Ω to V.

      Subset averages #

      theorem TauCeti.W1p.continuous_setAverage_value {E : Type u_1} [MeasurableSpace E] [NormedAddCommGroup E] [InnerProductSpace ℝ E] [BorelSpace E] {mu : MeasureTheory.Measure E} [mu.IsAddHaarMeasure] {U : TopologicalSpace.Opens E} {p : ENNReal} [Fact (1 ≤ p)] {S : Set E} (hSU : S ⊆ ↑U) (hSfin : mu S ≠ ⊤) :
      Continuous fun (v : ↥(W1p mu U p)) => ⨍ (y : E) in S, ↑↑(value v) y ∂mu

      The mean of the value of a first-order Sobolev function over a fixed subset of finite measure depends continuously on the function.