OAI/Analysis/LipschitzHilbert/Trees/SourcePhysicalGoodPath.lean

GitHub source file

5 declarations, 0 contain sorry, 0 declared as axioms

The weighted good-path construction for actual physical reference packets. A reference bundle only packages the original formulas; it has no probabilistic or path estimate as a field.

  1. structure OAI.LipschitzHilbert.PhysicalReferenceData line 15
  2. def OAI.LipschitzHilbert.PhysicalReferenceData.bad line 33
  3. theorem OAI.LipschitzHilbert.PhysicalReferenceData.bad_measurable line 37
  4. theorem OAI.LipschitzHilbert.PhysicalReferenceData.good_bounds line 41
  5. theorem OAI.LipschitzHilbert.eventually_physical_reference_good_path_area line 54

    Source documentation

    All shifts of each actual physical reference family enter one weighted good set. Its removed area depends on total top area, while a good path satisfies the three original quantitative reference bounds.

    theorem eventually_physical_reference_good_path_area (K D a : ℝ)
        (hK : 1 ≤ K) (hD : 2 < D) :
        ∀ᶠ L : ℝ in atTop, ∀ (ι σ : Type*) (tops : Finset ι) (T : ι → SlopeBox)
          (events : ι → Finset σ) (N : ℕ) (κ : Type) (Label : Fin N → Type)
          (Item : ∀ j, Label j → Type) [Fintype κ] [∀ j l, Fintype (Item j l)]
          (P : ι → σ → PhysicalReferenceData N κ Label Item)
          (A e s r : ℝ) (base : Set (ℝ×ℝ)),
          MeasurableSet base → (∀ t ∈ tops, ((events t).card:ℝ) ≤ L^a) →
          (∀ t ∈ tops, ∀ᵐ y : ℝ ∂volume, ∀ k : ℤ, ∀ i ∈ events t,
            (volume.restrict (sourceLineTile (T t).length k))
              {x | (x,y+(T t).slope*x) ∈ (P t i).bad L A e s r} ≤
            ENNReal.ofReal (Real.exp (-12*L))*
    …

    Mentions: OAI.LipschitzHilbert.SlopeBox, OAI.LipschitzHilbert.PhysicalReferenceData, OAI.LipschitzHilbert.sourceLineTile