Documentation

LeanPool.HighContrastHomogenization.Provider.Quenched.SmallContrastFusionIsotropy

High-contrast homogenization: Provider.Quenched.SmallContrastFusionIsotropy #

Imported from the Apache-2.0 HighContrastHomogenization development at commit 7a13dbcd8d6609264a713373f5c69ceeac870472.

The isotropy one-step chain and recursion constant steps 2 and 3, under the doctrine rule. #

The mapping of the fusion chain shows that almost none of it needs a parallel definition. weakDropConstant, weakBaseCoefficientSummed, weakSourceGroupSummed, split_at_recursionAlpha_at_level, iterationDropSum, per_generation_hrec_at_recursionAlpha_sharp_at_jb_src and the recursion-family core body are already abstract in the metric factor M, the load scale L and the recursion constant A; they carry no aspect ratio and are reused unchanged.

Π enters the one-step chain at exactly one link: the one-step-to-scalar conversion of HCPoly.Provider.Quenched.SmallContrastHrecInstance, whose hone carries rowValue2 Cd g mAl G s t kap and whose conclusion carries quadCoefficient d Cpre (rowCoefficient Cd g mAl G s t kap). Its proof is the row-value equation followed by one_step_scalar_shape, which is itself abstract in the row coefficient.

So the parallel copies needed are exactly three, and they are the ones here: one_step_to_scalar_step_isotropy, hrec_line_of_one_step_isotropy_sharp_at_jb_src and fusionRecursionConstantIsotropySharp. The last differs from fusionRecursionConstantIsotropySharp only in taking L as a parameter instead of fixing weakLoadScale Cd g mAl G kap in its body, and in using rowCoefficientIsotropy; fixing it in the body is precisely what cannot be repaired by an inequality.

The one-step scalar shape at the isotropy row #

theorem HCPolySupport.HighContrast.Quenched.one_step_to_scalar_step_isotropy (d : ℕ) {Cpre eta cRow kap wv W hatS hatT : ℝ} (hd0 : 0 ≤ ↑d) (hCpre : 0 ≤ Cpre) (hmaj : wv ≤ W ^ 2) (hone : hatT - 1 ≤ 4 * Cpre * ((3 / 2 + 1 / (4 * eta)) * (16 * ↑d * (hatS - hatT)) + rowValue2Isotropy d cRow kap (↑d * (hatT - 1)) + wv) + 2 * ((3 * ↑d + 4) * (↑d * (hatT - 1)) ^ 2)) :
↑d * (hatT - 1) ≤ dropCoefficient d Cpre eta * (↑d * (hatS - 1) - ↑d * (hatT - 1)) + quadCoefficient d Cpre (rowCoefficientIsotropy d cRow kap) * (↑d * (hatT - 1)) ^ 2 + weakCoefficient d Cpre * W ^ 2

The parallel copy of the one-step-to-scalar conversion. Identical proof: the row value is converted to its coefficient by an equation, and one_step_scalar_shape — abstract in the row coefficient — does the rest.

The recursion line at the isotropy row #

The isotropy recursion constant #

The weak-value base at an abstract load #

The fusion's binder body, Π-free #