Documentation

LeanPool.Superdiffusion.Algsuperdiff.Section4.Provider.ExcessDecay.BoundaryLaneWindows

BoundaryLaneWindows #

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

The truncated comparison windows of the boundary lane, and the choice y := x #

The §4.3 boundary regime works on the truncated windows

  W_j := (x + □_{n-j}) ∩ □_m ,        V := W_2 = (x + □_{n-2}) ∩ □_m ,

which are the windows of l.excess.decay.good.scales read through the replacement draft's item 2 (the lemma l.comparison.reduction). This module proves their geometry and the one statement the draft calls the window choice.

The window choice #

The printed proof asks for a point y ∈ □_m with both

  y + □_{n-2} ⊆ (z + □_n) ∩ □_m       and      (x + □_{n-2}) ∩ □_m ⊆ y + □_{n-2} ,

and the graph records the gap: "the existence of y with BOTH inclusions is asserted with no construction, and in the boundary case the two requirements pull against each other" — indeed the first inclusion is false at a boundary point x, because the full cube y + □_{n-2} then leaves □_m.

The draft's correction is y := x together with the move of the comparison function onto the truncated window: the pair of requirements becomes

  V = (x + □_{n-2}) ∩ □_m ⊆ (z + □_n) ∩ □_m       and      W_2 = V ,

and both are then immediate from x ∈ (z + □_{n-3}) ∩ □_m. That is truncatedWindow_subset_of_windowChoice and truncatedWindow_two_eq_V below; windowChoice_y_eq_x packages the pair.

What is not done here #

Nothing analytic: no harmonic function, no excess, no Caccioppoli estimate. In particular this module does not provide the volume-ratio (window sandwich) constant K_vol of the draft's §0 — only the qualitative positivity of the window volume, which is what the excess normalizers need to be finite.

References #

1. A general inclusion for translated origin cubes #

2. The truncated windows #

The truncated window (x + □_k) ∩ □_m of the §4.3 boundary regime, in the frozen theorem's own carrier shape (a translation image of an origin cube, intersected with the domain cube).

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

    The truncated windows are nested in the scale.

    The centre of a truncated window lies in it, as soon as it lies in the domain cube.

    3. The windows are open bounded convex domains #

    A truncated window centred in the domain cube has positive volume: it is a nonempty open set.

    4. The window choice y := x #

    The printed hypothesis x ∈ (z + □_j) ∩ □_m in its difference form.