OAI/Combinatorics/ContingencyTables/Dense/DenseGrid.lean

GitHub source file

10 declarations, 0 contain sorry, 0 declared as axioms

Conductance geometry for the actual dense penalty

The free table coordinates are reindexed by Fin q. The bin and cut-tube density comparisons below follow from the actual margins and integer widths. Thus the weighted cut theorem has no assumed penalty regularity or bin comparison when used for the dense completion sampler.

  1. def OAI.ContingencyTables.GridBoundary.realCorner line 25
  2. def OAI.ContingencyTables.GridBoundary.denseGridPenalty line 28
  3. theorem OAI.ContingencyTables.GridBoundary.denseGridPenalty_continuous line 32
  4. theorem OAI.ContingencyTables.GridBoundary.denseGridPenalty_convex line 37
  5. theorem OAI.ContingencyTables.GridBoundary.denseGridPenalty_change line 45
  6. theorem OAI.ContingencyTables.GridBoundary.closedBin_distance line 55
  7. theorem OAI.ContingencyTables.GridBoundary.tube_distances line 63
  8. theorem OAI.ContingencyTables.GridBoundary.denseGrid_bin_comparison line 102
  9. theorem OAI.ContingencyTables.GridBoundary.denseGrid_tube_comparison line 116
  10. theorem OAI.ContingencyTables.GridBoundary.denseGrid_weighted_cut line 138

    Source documentation

    Actual dyadic dense weights satisfy the interpolation cut bound, conditionally only on the published finite-box integral inequality.

    theorem denseGrid_weighted_cut (R : Option I → ℕ) (P : Option J → ℕ)
        (d : ℕ) (hd : 14 ≤ d) (hR : ∀ i, d ^ 12 ≤ R i) (hP : ∀ j, d ^ 12 ≤ P j)
        (hRmax : ∀ i, R (some i) ≤ R none) (hPmax : ∀ j, P (some j) ≤ P none)
        (hfree : Fintype.card (I × J) ≤ d) (e : I × J ≃ Fin q)
        (hPL : PublishedInputs.FiniteBoxPrekopaLeindler q)
        {K : ℕ} (Q : (Fin q → ℤ) → Prop) (t : ℝ) (ht : 0 < t) (ht₁ : t < 1)
        (hr : t * K ≤ 1 / 4) (hsize : (q - 1 : ℕ) * (2 * (t * K)) ≤ (1 / 2 : ℝ))
        (hSmass : 0 < ∫ x in selectedBins K Q,
          DenseWeights.density ((d : ℝ) ^ 4) (denseGridPenalty R P d e) x)
        (hcut : 2 * (∫ x in selectedBins K Q,
          DenseWeights.density ((d : ℝ) ^ 4) (denseGridPenalty R P d e) x) ≤
          ∫ x in {x | Cube K x},
    …

    Mentions: OAI.ContingencyTables.PublishedInputs.FiniteBoxPrekopaLeindler, OAI.ContingencyTables.GridBoundary.selectedBins, OAI.ContingencyTables.DenseWeights.density, OAI.ContingencyTables.GridBoundary.denseGridPenalty, OAI.ContingencyTables.GridBoundary.Cube