Documentation

LeanPool.Superdiffusion.Algsuperdiff.Section5.Provider.StoppedMomentsCubeModel

StoppedMomentsCubeModel #

Ported from scottnarmstrong/Superdiffusion at 7f20476385d4af418f110303085ad0740c9eba46. Module imports, support namespace and local option directives adapted for Lean Pool.

The stopped moments of the stream process on a triadic cube, at the model #

The stopped-moment bounds of the displacement estimates are composed here at the model: the process is the stream process of a full sample, the cube is the triadic cube centred at the origin at a scale m, the observables are the linear and quadratic data cut off by a smooth function equal to one on the closed cube, and the corrector bounds are the ones of the renormalization theorem at the sample.

The objects the quenched moment composition takes as free parameters are defined:

and the two stopped bounds are proved for them, in the exact form that composition consumes:

Both are proved under the corrector clause of the renormalization theorem at the sample and the scale m, with the effective diffusivity sigmaBar and the error value EB of that clause; the clause holds almost surely at every scale for the families of exists_renormalizationFamily_corrector, and the almost-sure form on the full-sample carrier is recorded at the end. The side condition d sigmaBar t ≤ (3^m)² of the composition is derived from the defining property of the confinement scale.

Main definitions #

Main results #

References #

1. The cut-off of the cube #

theorem Algsuperdiff.Section5.Provider.exists_cubeCutoff (d : ℕ) (m : ℤ) :
∃ (c : HCPolySupport.Vec d → ℝ), ContDiff ℝ (↑⊤) c ∧ (∀ (x : HCPolySupport.Vec d), 0 ≤ c x ∧ c x ≤ 1) ∧ (∀ x ∈ closure (Support.cubeSetAt 0 m), c x = 1) ∧ tsupport c ⊆ Support.cubeSetAt 0 (m + 1)

A smooth cut-off between 0 and 1, equal to one on the closed cube of scale m centred at the origin and supported in the open cube of the next scale.

The cut-off of the cube of scale m centred at the origin: smooth, between 0 and 1, equal to one on the closed cube, supported in the open cube of the next scale.

Equations
Instances For

    2. The bounds for the cut-off observables #

    A point of the cube of scale m + 1 has supremum norm below half its side.

    The cut-off linear observables are uniformly bounded, in every direction.

    The cut-off quadratic observables are uniformly bounded, in every direction.

    3. The basis directions #

    4. The process quantities #

    The exit probability: the probability that the stream process started at the origin has left the cube of scale m centred at the origin by the time t.

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

      The stopped mean vector: its i-th coordinate is the mean, under the stream process started at the origin, of the cut-off linear observable in the direction i at the position of the path at its exit time from the cube of scale m centred at the origin, truncated at t.

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

        5. The squared stopped mean #

        The squared norm of the stopped mean vector. Each coordinate is the stopped mean of the cut-off linear observable in that direction, which is within 2 · 3^m EB of zero by the stopped mean bound and the corrector estimate at the linear datum; summing the squares over the d directions gives 4 d (3^m)² EB².

        6. The stopped second moment #

        theorem Algsuperdiff.Section5.Provider.abs_cubeStoppedQuadratic_sub_le {d : ℕ} [NeZero d] (M : Section3.ABKModel d) (omega : Field.FullSample d M.gamma) (m : ℤ) {sigmaBarM EBm : ℝ} (hsigma : 0 < sigmaBarM) (hEB : 0 ≤ EBm) (hren : ∀ (L : ℤ), m ≤ L → ∀ (u v h : HCPolySupport.H1Function (HCPolySupport.openCubeSet (HCPolySupport.originCube d m))) (g : HCPolySupport.Vec d → HCPolySupport.Vec d) (Kg Kh KhInf : ℝ), Section4.Support.IsDirichletSolutionOn (Section3.Cutoff.coefficientCutoff M.nu L ↑omega).toCoeffField (HCPolySupport.originCube d m) u h g → Section4.Support.IsDirichletSolutionOn (fun (x : HCPolySupport.Vec d) => sigmaBarM • 1) (HCPolySupport.originCube d m) v h g → Section4.Support.HolderSeminormBoundOn (HCPolySupport.openCubeSet (HCPolySupport.originCube d m)) (1 / 2) Kg g → Section4.Support.HolderSeminormBoundOn (HCPolySupport.openCubeSet (HCPolySupport.originCube d m)) (1 / 2) Kh h.grad → (∀ x ∈ HCPolySupport.openCubeSet (HCPolySupport.originCube d m), ‖h.grad x‖ ≤ KhInf) → Section4.Support.HasGradientOn (HCPolySupport.openCubeSet (HCPolySupport.originCube d m)) h.toFun h.grad → ∀ᵐ (x : HCPolySupport.Vec d) ∂MeasureTheory.volume.restrict (HCPolySupport.openCubeSet (HCPolySupport.originCube d m)), Real.rpow 3 (-↑m) * |u.toFun x - v.toFun x| ≤ EBm * (sigmaBarM⁻¹ * Real.rpow 3 (↑m / 2) * Kg + (KhInf + Real.rpow 3 (↑m / 2) * Kh))) (t : NNReal) (i : Fin d) :

        The stopped second moment in every direction. The stopped mean of the cut-off quadratic observable in the direction i is within 10 (3^m)² EB plus 2 sigmaBar t times the exit probability of 2 sigmaBar t, by the stopped second-moment bound with the corrected observable and the corrector estimate at the quadratic datum.

        7. The side condition of the composition #

        theorem Algsuperdiff.Section5.Provider.dim_mul_sigmaBar_mul_le_three_zpow_sq_of_isConfinementScale {d : ℕ} {nu cstar gamma t K : ℝ} (hnu : 0 < nu) (hcstar : 0 < cstar) (hgamma : 0 < gamma) (hgamma4 : gamma ≤ 1 / 4) (ht : 0 < t) (hK : 1 ≤ K) (hKd : 2 * ↑d ≤ K ^ 2) {Stilde : ℤ → ENNReal} {m : ℤ} (hm : Support.IsConfinementScale Stilde (K * √|Real.log gamma| * Support.intrinsicScale nu cstar gamma t) m) {sigmaBar C : ℝ} (hsigma : 0 < sigmaBar) (hC : 0 ≤ C) (hsmall : C * √gamma * |Real.log gamma| ≤ 1 / 2) (hsmall' : ↑d * (1 + 2 * C) * √gamma * |Real.log gamma| ≤ 1 / 2) (hprofile : |sigmaBar - Support.effectiveDiffusivity nu cstar gamma (3 ^ m)| ≤ C * √gamma * |Real.log gamma| * sigmaBar) :
        ↑d * (sigmaBar * t) ≤ (3 ^ m) ^ 2

        The confinement scale dominates the diffusive scale. At a confinement scale m for the length K |log γ|^{1/2} R(t), with K² ≥ 2 d, the running diffusivity satisfies d sigmaBar t ≤ (3^m)², once the disorder parameter is small enough that d (1 + 2 C) γ^{1/2} |log γ| ≤ 1/2.

        8. The corrector clause on the full-sample carrier #

        theorem Algsuperdiff.Section5.Provider.ae_renormalizationCorrector_fullSample {d : ℕ} (M : Section3.ABKModel d) {sigmaBar : ℤ → ℝ} {EB : ℤ → Section3.Cutoff.CutoffSample d → ℝ} (hdir : ∀ (m : ℤ), ∀ᵐ (omega : Section3.Cutoff.CutoffSample d) ∂↑(Section3.Cutoff.cutoffSampleLaw M), ∀ (L : ℤ), m ≤ L → ∀ (u v h : HCPolySupport.H1Function (HCPolySupport.openCubeSet (HCPolySupport.originCube d m))) (g : HCPolySupport.Vec d → HCPolySupport.Vec d) (Kg Kh KhInf : ℝ), Section4.Support.IsDirichletSolutionOn (Section3.Cutoff.coefficientCutoff M.nu L omega).toCoeffField (HCPolySupport.originCube d m) u h g → Section4.Support.IsDirichletSolutionOn (fun (x : HCPolySupport.Vec d) => sigmaBar m • 1) (HCPolySupport.originCube d m) v h g → Section4.Support.HolderSeminormBoundOn (HCPolySupport.openCubeSet (HCPolySupport.originCube d m)) (1 / 2) Kg g → Section4.Support.HolderSeminormBoundOn (HCPolySupport.openCubeSet (HCPolySupport.originCube d m)) (1 / 2) Kh h.grad → (∀ x ∈ HCPolySupport.openCubeSet (HCPolySupport.originCube d m), ‖h.grad x‖ ≤ KhInf) → Section4.Support.HasGradientOn (HCPolySupport.openCubeSet (HCPolySupport.originCube d m)) h.toFun h.grad → ∀ᵐ (x : HCPolySupport.Vec d) ∂MeasureTheory.volume.restrict (HCPolySupport.openCubeSet (HCPolySupport.originCube d m)), Real.rpow 3 (-↑m) * |u.toFun x - v.toFun x| ≤ EB m omega * ((sigmaBar m)⁻¹ * Real.rpow 3 (↑m / 2) * Kg + (KhInf + Real.rpow 3 (↑m / 2) * Kh))) :

        The corrector clause holds almost surely on the full-sample carrier, at every scale. The clause of the renormalization families is almost sure on the cutoff-sample carrier at each scale; countability of the scales and the identification of the full-sample law with the cutoff-sample law transport it.