Documentation

Atlas.DifferentialAnalysis.code.SmoothingOperators

Smoothing by a tempered distribution: convolving a tempered distribution u with a Schwartz function φ (via the family of translations x ↦ u(φ(·−x))) produces a smooth function.

The Fréchet derivative of z ↦ L(ψ(·−z)) at x evaluated on h equals L(∂_{-h} ψ (·−x)): differentiating the family of translations in the spatial variable produces a line derivative of ψ (in the opposite direction) inside L.

Line derivative of the smoothed function: the directional derivative of x ↦ u(φ(·−x)) along m equals u applied to ∂_{-m} φ translated to x. This is the distributional identity used to differentiate the smoothing.

noncomputable def DifferentialOperators.evalAtReal {n : } (P : MvPolynomial (Fin n) ) (ξ : EuclideanSpace (Fin n)) :

Evaluation of a multivariate complex polynomial at a real Euclidean point, viewed as a function EuclideanSpace ℝ (Fin n) → ℂ.

Instances For

    The polynomial P has a polynomial lower bound of degree m with constant C > 0 if, for every ξ of norm larger than 1/C, the inequality ‖P(ξ)‖ ≥ C · ‖ξ‖^m holds.

    Instances For

      The pointwise reciprocal of a polynomial: ξ ↦ 1 / P(ξ), with P(ξ) = 0 mapped to 0 by the inverse convention.

      Instances For