Documentation

LeanPool.OSforGFF.Covariance.RealForm

Real Covariance Form and Square Root Propagator Embedding #

This file provides the real covariance form infrastructure along with the proof of the square root propagator embedding theorem.

Main Results #

Real Covariance Form #

noncomputable def QFT.freeCovarianceFormR (m : ) (f g : TestFunction) :

Real covariance bilinear form induced by the free covariance kernel.

Equations
Instances For

    Weighted L² Space Construction #

    The weighted measure on momentum space with density (‖k‖² + m²)⁻¹.

    Equations
    Instances For

      ℝ-linear view of the complex Fourier transform on Schwartz space.

      Note: we build this manually instead of using restrictScalars because the LinearMap.CompatibleSMul instance is not found automatically in mathlib v4.29.

      Equations
      • One or more equations did not get rendered due to their size.
      Instances For

        ℝ-linear view of the Schwartz-to- embedding.

        Equations
        • One or more equations did not get rendered due to their size.
        Instances For

          The Embedding Map #

          noncomputable def QFT.sqrtPropagatorMap (m : ) (f : TestFunction) :

          The embedding T maps a test function to a weighted function in momentum space. Conceptually: T f = FourierTransform(f) * (‖k‖² + m²)^(-1/2).

          Equations
          Instances For

            The sqrtPropagatorMap is square-integrable.

            The weighted Fourier representative lies in L².

            noncomputable def QFT.sqrtPropagatorMapNormSq (m : ) (f : TestFunction) :

            The squared L² norm of the mapped function.

            Equations
            Instances For

              The map is linear in f (additive).

              The map is ℝ-linear (scalar multiplication).

              Connection to Covariance #

              For real test functions, the star (conjugation) of toComplex is the identity.

              For real test functions, freeCovarianceℂ agrees with freeCovarianceℂBilinear.

              Key lemma: The squared norm equals the covariance form.

              The Proof of sqrtPropagatorEmbedding #

              @[reducible, inline]

              The target Hilbert space: L²(SpaceTime, Lebesgue) with complex values.

              Equations
              Instances For

                The linear map T: TestFunction → L².

                Equations
                Instances For

                  Squared L² norm of the embedded function in terms of the pointwise integral.

                  theorem QFT.sqrtPropagatorEmbedding (m : ) [Fact (0 < m)] :

                  Existence of a linear embedding realizing the free covariance as a squared norm. The target space H is an inner product space (L² is a Hilbert space). Note: InnerProductSpace ℝ H implies NormedSpace ℝ H via InnerProductSpace.toNormedSpace.

                  Auxiliary Lemmas for Continuity #

                  The embedding map TestFunction → L² is continuous.

                  Continuity of the real covariance quadratic form f ↦ C(f,f).

                  Positivity and Other Properties #

                  Positivity of the real covariance quadratic form.

                  Symmetry of the real covariance bilinear form.

                  theorem QFT.freeCovarianceFormR_add_left (m : ) [Fact (0 < m)] (f₁ f₂ g : TestFunction) :

                  Linearity in the first argument of the real covariance bilinear form.

                  Scalar multiplication in the first argument of the real covariance bilinear form.

                  theorem QFT.freeCovarianceFormR_add_right (m : ) [Fact (0 < m)] (f g₁ g₂ : TestFunction) :

                  Addition in the second argument of the real covariance bilinear form.

                  Scalar multiplication in the second argument of the real covariance bilinear form.

                  Zero in the first argument gives zero.

                  Zero in the second argument gives zero.

                  Mixed-time-reflection identity for the real free covariance.

                  theorem QFT.freeCovarianceFormR_left_linear_any_right (m : ) [Fact (0 < m)] {n : } (f : Fin nPositiveTimeTestFunction) (c : Fin n) (s : Finset (Fin n)) (g : TestFunction) :
                  is, c i * freeCovarianceFormR m (compTimeReflectionReal (f i)) g = freeCovarianceFormR m (∑ is, c i compTimeReflectionReal (f i)) g

                  Left linearity of freeCovarianceFormR for any fixed right argument.