Documentation

LeanPool.HighContrastHomogenization.Entry.Response.Rows.CutoffMeanRowCarriers

High-contrast homogenization: Entry.Response.Rows.CutoffMeanRowCarriers #

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

The cutoff-mean row on the response's own data #

The cutoff-mean row of p.response.transfer is the sum of a cell half, bounded from the response's own cell data by (L_s^∓ 2 τ^∓)^{1/2}, and an oscillation half, read at the dual variable Y^∓ and bounded by its depth-H and descendant contributions against c₂ 3^{-H}(E[J_t^∓] L_s^∓)^{1/2}; this file combines the two into the row's final concrete form for both signs. The oscillation bound is stated through a single constant, here shown to be nonnegative and to satisfy the elementary arithmetic identity the combination needs, and a short identity for the cutoff's own fluctuation average closes the remaining bookkeeping. Together these give the cutoff-mean row exactly in the form the cutoff estimate consumes, with no further hypothesis on the coefficient sample.

The cutoff-mean row at the carriers #

The cutoff-mean row of p.response.transfer is the sum of a cell half and an oscillation half. CellHalfFinalBound bounds the cell half by (L_s^∓ 2 tau^∓)^{1/2} from the carriers' own annealed data. The oscillation half of CutoffOscillationHalfClosure, read at the dual variable Y^∓, bounds the scalar annealed functional ⨍_{U_t}(φ-1)(⟨Y₂, ∇u⟩ + ⟨Y₁, a∇u⟩) against its depth-H cell part; CellHalfFinalBound identifies the crossed pairing of the vector oscillation part with exactly that difference, once the within-cell fluctuation is split by volumeAverage_fluct_eq_osc_add_cell. Together with the coordinate integrabilities of CutoffMeanDefectIntegrability, the row assembly of OptimizerMeanRowCarriers then gives the named obligation CutoffMeanRow∓ with no sample-side premise left.

The constant of the oscillation half of the cutoff-mean row: the cutoff Lipschitz constant 32 d² Θ, the square root of the geometric series ∑ 3^{-n/2} of the descendant weights, the generation shift of the source-load layers and the √8 of the doubled cell energy.

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

    The oscillation-half constant is nonnegative.

    theorem HCPolySupport.HighContrast.Multiscale.cutoffMeanRowMinus_of_carriers {d : ℕ} [NeZero d] (hd : 2 ≤ d) (P : MeasureTheory.Measure (CoeffSpace d)) [MeasureTheory.IsProbabilityMeasure P] (hstat : IsStationaryLaw P) (γ : ℝ) (E : BlockMat d) (Ψ : ℝ → ℝ) (Kg : ℝ) (Src : CoeffSpace d → ℝ) (hdag : CoarseEllipticityDagger P γ E Ψ Kg Src) (jStar : ℕ) (hjStar : 2 * d ≤ 3 ^ jStar) (F : BlockMat d) (hm : Matrix.PosDef (explicitCanonicalMetric F)) (hF : Matrix.PosDef (toFullBlockMat F)) (H : ℕ) (s t : ℤ) (ht : t = s + ↑H) (hjs : ↑jStar ≤ s) (e : Vec d) (φ : Vec d → ℝ) (hφ : IsResponseCutoff (respGrid jStar F) t φ) (uM : (a : CoeffSpace d) → AHarmonicFunction (respCoeffMinus F a) (respCell jStar F t)) (hmax : ∀ (a : CoeffSpace d), IsResponseMaximizer (respCell jStar F t) (respP (respMean P jStar F t) e) (respqMinus P jStar F t e) (respCoeffMinus F a) (uM a)) (hblkw : ∀ (k : ℤ) (w : Fin d → ℤ), HasIntegrableCoarseBlock P (adaptedCellAtCenter (respGrid jStar F) k w)) (hblkt : HasIntegrableCoarseBlock P (respCell jStar F t)) (hJt : MeasureTheory.Integrable (fun (a : CoeffSpace d) => respJ (respGrid jStar F) t (respP (respMean P jStar F t) e) (respqMinus P jStar F t e) (respCoeffMinus F a)) P) (hJs : MeasureTheory.Integrable (fun (a : CoeffSpace d) => respJ (respGrid jStar F) s (respP (respMean P jStar F t) e) (respqMinus P jStar F t e) (respCoeffMinus F a)) P) (hsumm : ∀ (Y : BlockVec d), Summable (respSourceLoadSummand P jStar F s (respCoeffMinus F) Y)) (hτ : 0 ≤ respTauMinus P jStar F s t e) :
    CutoffMeanRowMinus (max 1 (respCutoffOscConst d)) P jStar F H s t e φ uM

    The cutoff-mean row at the carriers, minus sign. Under the annealed data of the response window — stationarity, coarse ellipticity, the invertible rounded grid, the maximizer family, the two terminal responses and the convergence of the source-load series at every dual vector — the named obligation CutoffMeanRowMinus holds with constant max 1 (respCutoffOscConst d).

    theorem HCPolySupport.HighContrast.Multiscale.cutoffMeanRowPlus_of_carriers {d : ℕ} [NeZero d] (hd : 2 ≤ d) (P : MeasureTheory.Measure (CoeffSpace d)) [MeasureTheory.IsProbabilityMeasure P] (hstat : IsStationaryLaw P) (γ : ℝ) (E : BlockMat d) (Ψ : ℝ → ℝ) (Kg : ℝ) (Src : CoeffSpace d → ℝ) (hdag : CoarseEllipticityDagger P γ E Ψ Kg Src) (jStar : ℕ) (hjStar : 2 * d ≤ 3 ^ jStar) (F : BlockMat d) (hm : Matrix.PosDef (explicitCanonicalMetric F)) (hF : Matrix.PosDef (toFullBlockMat F)) (H : ℕ) (s t : ℤ) (ht : t = s + ↑H) (hjs : ↑jStar ≤ s) (e : Vec d) (φ : Vec d → ℝ) (hφ : IsResponseCutoff (respGrid jStar F) t φ) (uP : (a : CoeffSpace d) → AHarmonicFunction (respCoeffPlus F a) (respCell jStar F t)) (hmax : ∀ (a : CoeffSpace d), IsResponseMaximizer (respCell jStar F t) (respP (respMean P jStar F t) e) (respqPlus P jStar F t e) (respCoeffPlus F a) (uP a)) (hblkw : ∀ (k : ℤ) (w : Fin d → ℤ), HasIntegrableCoarseBlock P (adaptedCellAtCenter (respGrid jStar F) k w)) (hblkt : HasIntegrableCoarseBlock P (respCell jStar F t)) (hJt : MeasureTheory.Integrable (fun (a : CoeffSpace d) => respJ (respGrid jStar F) t (respP (respMean P jStar F t) e) (respqPlus P jStar F t e) (respCoeffPlus F a)) P) (hJs : MeasureTheory.Integrable (fun (a : CoeffSpace d) => respJ (respGrid jStar F) s (respP (respMean P jStar F t) e) (respqPlus P jStar F t e) (respCoeffPlus F a)) P) (hsumm : ∀ (Y : BlockVec d), Summable (respSourceLoadSummand P jStar F s (respCoeffPlus F) Y)) (hτ : 0 ≤ respTauPlus P jStar F s t e) :
    CutoffMeanRowPlus (max 1 (respCutoffOscConst d)) P jStar F H s t e φ uP

    The cutoff-mean row at the carriers, minus sign. Under the annealed data of the response window — stationarity, coarse ellipticity, the invertible rounded grid, the maximizer family, the two terminal responses and the convergence of the source-load series at every dual vector — the named obligation CutoffMeanRowPlus holds with constant max 1 (respCutoffOscConst d).