Coarse-graining support: Support.Book.Ch05.Theorems.Section51.AnnealedConvergence #
Imported from the Apache-2.0 CoarseGraining development at commit
c7ddd76c08ade64fed1b8d2ca51be14dfee8deb4.
Theorem t.annealed.convergence #
The main theorem combines the perturbative entry mechanism with the
small-contrast algebraic iteration, applied after shifting the (P4) exponents
by the Section 5.3 fluctuation buffer.
The first coordinate unit vector, used to test annealed response bounds in a fixed unit direction.
Instances For
theorem
HCPolySupport.Book.Ch05.Section51.annealedConvergence_homogenizationScale
{d : ℕ}
[NeZero d]
(params : QuantitativeCoarseGrainedEllipticityParams d)
:
∃ (C : ℝ) (α : ℝ),
0 < C ∧ 0 < α ∧ ∀ {P : Ch04.RestrictionCoeffLaw d} (hP : Ch04.RestrictionLawCarrier P) (hStruct : Ch04.RestrictionStructuralLaw P)
(hP4 : QuantitativeCoarseGrainedEllipticity P),
hP4.params = params →
∀ (n : ℕ), thetaAtScale hP hStruct ↑(annealedAlgebraicEntryScale P hP4 C + n) ≤ 1 + Real.rpow 3 (-α * ↑n)
Theorem t.annealed.convergence: convergence of the annealed contrast.
The constants are selected from the quantitative ellipticity parameter record before the probability law, hence are independent of the law itself.