Documentation

Atlas.DifferentialAnalysis.code.TestFunctions

A continuous, complex-valued function on ℝⁿ that vanishes at infinity, is , and whose first-order partial derivatives also vanish at infinity.

Instances For
    noncomputable def TestFunctions.restrictDual {𝕜 : Type u_1} [NontriviallyNormedField 𝕜] {V : Type u_2} [SeminormedAddCommGroup V] [NormedSpace 𝕜 V] {U : Type u_3} [SeminormedAddCommGroup U] [NormedSpace 𝕜 U] (ι : V →L[𝕜] U) :
    (U →L[𝕜] 𝕜) →L[𝕜] V →L[𝕜] 𝕜

    Pullback of continuous linear functionals along a continuous linear map ι : V →L[𝕜] U: sends a dual element on U to its restriction along ι.

    Instances For
      @[simp]
      theorem TestFunctions.restrictDual_apply {𝕜 : Type u_1} [NontriviallyNormedField 𝕜] {V : Type u_2} [SeminormedAddCommGroup V] [NormedSpace 𝕜 V] {U : Type u_3} [SeminormedAddCommGroup U] [NormedSpace 𝕜 U] (ι : V →L[𝕜] U) (L : U →L[𝕜] 𝕜) :
      (restrictDual ι) L = L.comp ι

      restrictDual ι L is just L ∘ ι.

      theorem TestFunctions.norm_restrictDual_le {𝕜 : Type u_1} [NontriviallyNormedField 𝕜] {V : Type u_2} [SeminormedAddCommGroup V] [NormedSpace 𝕜 V] {U : Type u_3} [SeminormedAddCommGroup U] [NormedSpace 𝕜 U] (ι : V →L[𝕜] U) (L : U →L[𝕜] 𝕜) {C : } (hC : ι C) :

      Operator-norm bound for restrictDual: if ‖ι‖ ≤ C, then ‖L ∘ ι‖ ≤ C * ‖L‖.

      theorem TestFunctions.restrictDual_injective {𝕜 : Type u_1} [NontriviallyNormedField 𝕜] {V : Type u_2} [SeminormedAddCommGroup V] [NormedSpace 𝕜 V] {U : Type u_3} [SeminormedAddCommGroup U] [NormedSpace 𝕜 U] (ι : V →L[𝕜] U) (hd : DenseRange ι) :

      If ι has dense range, then its dual restrictDual ι is injective: a continuous functional is determined by its values on a dense subspace.

      theorem TestFunctions.restrictDual_norm_bound_and_injective {𝕜 : Type u_1} [NontriviallyNormedField 𝕜] {V : Type u_2} [SeminormedAddCommGroup V] [NormedSpace 𝕜 V] {U : Type u_3} [SeminormedAddCommGroup U] [NormedSpace 𝕜 U] (ι : V →L[𝕜] U) (C : ) (hC : ι C) :
      (∀ (L : U →L[𝕜] 𝕜), (restrictDual ι) L C * L) (DenseRange ιFunction.Injective (restrictDual ι))

      Combined statement: restrictDual is bounded by C on operator norms, and is injective whenever ι has dense range.

      def TestFunctions.ckNorm (n a✝ : ) :
      (EuclideanSpace (Fin n))

      The C^k sup-norm of a function f : ℝⁿ → ℂ: sum of sup norms of all partial derivatives up to order k.

      Instances For

        The space of C^k functions on ℝⁿ vanishing at infinity together with all of their iterated derivatives up to order k.

        Instances For

          The C^k-norm of an element of ContDiffZeroAtInftyN n k.

          Instances For
            noncomputable def TestFunctions.japaneseBracket (n : ) (x : EuclideanSpace (Fin n)) :

            The Japanese bracket ⟨x⟩ := √(1 + ‖x‖²), a smooth weight comparable to ‖x‖ at infinity used to define weighted function spaces.

            Instances For

              The Japanese bracket ⟨x⟩ is strictly positive.

              The Japanese bracket ⟨x⟩ is nonzero.

              Weighted C^k-functions on ℝⁿ vanishing at infinity: an element is represented by a witness v ∈ C^k whose ratio against ⟨x⟩^{-l} lies in the original space.

              Instances For

                Underlying function ⟨x⟩^{-l} v(x) for a weighted C^k element with witness v.

                Instances For

                  The Schwartz space described as functions on ℝⁿ lying in every weighted C^k-space with weight ⟨x⟩^{-l} (Melrose Prop 7.3 characterisation).

                  Instances For
                    @[reducible, inline]

                    Standard Schwartz space 𝓢(ℝⁿ, ℂ) abbreviation.

                    Instances For
                      @[implicit_reducible]

                      SchwartzTestFunctionSpace n coerces to functions ℝⁿ → ℂ.

                      theorem TestFunctions.SchwartzTestFunctionSpace.ext {n : } {f g : SchwartzTestFunctionSpace n} (h : ∀ (x : EuclideanSpace (Fin n)), f x = g x) :
                      f = g

                      Function-extensionality for elements of SchwartzTestFunctionSpace.

                      @[simp]
                      theorem TestFunctions.SchwartzTestFunctionSpace.coe_mk {n : } (f : EuclideanSpace (Fin n)) (hf : ∀ (k l : ), ∃ (w : WeightedContDiffZeroAtInfty n k l), ∀ (x : EuclideanSpace (Fin n)), f x = WeightedContDiffZeroAtInfty.toFun n w x) :
                      { toFun := f, mem_weightedSpace := hf } = f

                      The coercion of mk f hf is just f.

                      Equivalence between our concrete SchwartzTestFunctionSpace n and Mathlib's SchwartzMap (ℝⁿ, ℂ), expressing Melrose's characterisation of Schwartz space.

                      Instances For

                        Each coordinate of a vector in ℝⁿ is bounded in absolute value by the Euclidean norm of the vector.

                        The operator norm of a continuous linear functional T : ℝⁿ →L[ℝ] ℂ is bounded by the sum of its values on the standard basis.