Documentation

LeanPool.Superdiffusion.Algsuperdiff.Section3.Provider.ErrorComparison.ExponentMonotonicity

ExponentMonotonicity #

Ported from scottnarmstrong/Superdiffusion at 7f20476385d4af418f110303085ad0740c9eba46. Module imports, support namespace and local option directives adapted for Lean Pool.

Provider: the ordered multiscale ellipticity gauges #

This file proves a local version of the display e.ellipticities.monotone.ordered of ABK26:

For every m ∈ ℤ, q ∈ [1,∞] and t, s ∈ (0,∞) with t < s, λ_{t,q}(□_m; a) ≤ λ_{s,q}(□_m; a) ≤ |σ_*^{-1}(□_m; a)|^{-1} ≤ |b(□_m; a)| ≤ Λ_{s,q}(□_m; a) ≤ Λ_{t,q}(□_m; a).

The carriers are CoarseGraining's Chapter 2.5 objects: Book.Ch02.lambdaSq and Book.Ch02.LambdaSq for the two multiscale gauges (definition e.coarse.grained.ellipticity / e.coarse.grained.ellipticity.infty, i.e. ABK26 (2.22)--(2.23)), and Book.Ch02.coarseSigmaStarInvMatrixNorm, Book.Ch02.coarseBMatrixNorm for the two one-cube norms |σ_*^{-1}(U; a)| and |b(U; a)| of e.CG.bounds.1.

Scope and carrier notes (implementation record, no manuscript defect) #

Proof route #

Nothing analytic is reproved.

Main results #

References #

The display #

e.ellipticities.monotone.ordered (ABK26). For every triadic cube, every admissible exponent q ∈ [1,∞] and all 0 < t < s, the six multiscale and one-cube ellipticity gauges are ordered:

λ_{t,q} ≤ λ_{s,q} ≤ |σ_*^{-1}|^{-1} ≤ |b| ≤ Λ_{s,q} ≤ Λ_{t,q}.

Consumer-facing corollaries #

Cube monotonicity #

Cube monotonicity of the multiscale lower ellipticity. If R is a descendant of Q at scale k, then

λ_{s,q}(Q; a) ≤ 3^{2s(Q.scale − k)} λ_{s,q}(R; a).