Documentation

LeanPool.HighContrastHomogenization.Entry.Multiscale.Global.OutputAndRun

High-contrast homogenization: Entry.Multiscale.Global.OutputAndRun #

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

The selected output and the run #

The profile bound satisfied by the selected output, the two inputs (p.scale.selection and p.initial.fixed.grid.scale), the predicate SelectedOutput describing the tuple the run produces, and the run itself.

§F Selected-output estimates (p.global.selection) #

theorem HCPolySupport.HighContrast.Multiscale.output_profile_bound (Cout Cfs Cmaj Cexp Q ρ σ H d pss dss pts dt ptt Δ : ℝ) (hρ : 0 ≤ ρ) (hCout : 0 ≤ Cout) (hCfs : 0 ≤ Cfs) (hCmaj : 0 ≤ Cmaj) (hCexp : 0 ≤ Cexp) (hH : 0 ≤ H) (hQ : 0 ≤ Q) (hd : 0 ≤ d) (hσ : 0 ≤ σ) (hpss : 0 ≤ pss) (hdss : 0 ≤ dss) (hpts : 0 ≤ pts) (hdt : 0 ≤ dt) (hptt : 0 ≤ ptt) (hΔ0 : 0 ≤ Δ) (hΔ : Δ < d * σ) (h1 : pss + dss ≤ Cout * ρ) (h2 : pts + dt ≤ Cfs * H * (pss + dss + (Real.exp (Q * Δ) - 1))) (hexp : Real.exp (Q * d * σ) - 1 ≤ Cexp * ρ) (h3 : ptt ≤ Cmaj * pts) :
max (max pss pts) ptt + dss + dt ≤ (2 * Cout + (2 + Cmaj) * Cfs * H * (Cout + Cexp)) * ρ

p.global.selection (e.global.selection.profile), scalar assembly with ρ = σ^((1-γ)/8): change-of-geometry output pss + dss ≤ Cout ρ, fixed-span s → t with Δ_{s,t} < dσ, the exponential bound exp(Qdσ) - 1 ≤ Cexp ρ, majorization ptt ≤ Cmaj pts. The resulting Cprof = 2 Cout + (2 + Cmaj) Cfs H (Cout + Cexp) is free of σ.

§G The proposition inputs and the run assembly #

theorem HCPolySupport.HighContrast.Multiscale.initial_provider_input (d : ℕ) (hd : 2 ≤ d) :
∃ (Cgeom : ℝ), 0 < Cgeom ∧ ∀ γ ∈ Set.Ico 0 1, ∀ (S : SelectionData), S.Selects d γ → ∃ (Csrc : ℝ), 0 < Csrc ∧ ∀ ηinit ∈ Set.Ioc 0 1, ∃ (C : ℝ), 0 < C ∧ ∀ (ε σ : ℝ), 0 < σ → σ ≤ ε → ε ≤ S.eps0 → ∀ (P : MeasureTheory.Measure (CoeffSpace d)) (E : BlockMat d) (Ψ : ℝ → ℝ) (K : ℝ) (Src : CoeffSpace d → ℝ), MeasureTheory.IsProbabilityMeasure P → IsStationaryLaw P → IsUnitRangeLaw P → CoarseEllipticityDagger P γ E Ψ K Src → ∀ (jStar : ℕ), 2 * d ≤ 3 ^ jStar → ⌈Csrc * Real.logb 3 (2 * K)⌉ ≤ ↑jStar → ∀ (B : ℝ), S.B0 ε σ ≤ B → ∃ (n₀ : ℤ), ↑jStar + ⌈B * Real.logb 3 (2 + aspectRatio E)⌉ ≤ n₀ ∧ n₀ ≤ ↑jStar + ⌈B * Real.logb 3 (2 + aspectRatio E)⌉ + ⌈C * Real.logb 3 (2 + aspectRatio E)⌉ ∧ profile P γ 1 jStar n₀ n₀ + determinantDrift P γ 1 jStar n₀ ≤ ηinit ∧ BlockMatLoewnerLE (blockScale (1 / 2) (ofFullBlockMat (toFullBlockMat (blockSwap d) * (toFullBlockMat E)⁻¹ * toFullBlockMat (blockSwap d)))) (adaptedMean P 1 n₀) ∧ BlockMatLoewnerLE (adaptedMean P 1 n₀) (blockScale 2 E) ∧ projectiveDistance 1 (explicitCanonicalMetric (adaptedMean P 1 n₀)) ≤ Cgeom * Real.log (2 + 4 * aspectRatio E)

p.initial.fixed.grid.scale, exact type declared in HCPoly/Entry/Statements/InitialFixedGridScale.lean; applies Entry.initial_fixed_grid_scale directly.

def HCPolySupport.HighContrast.Multiscale.SelectedOutput {d : ℕ} (P : MeasureTheory.Measure (CoeffSpace d)) (γ ε σ Cprof C : ℝ) (E : BlockMat d) (H jStar : ℕ) (B : ℝ) :

The output tuple of the run (p.global.selection), as a predicate, so global_run and global_selection share one interface.

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