Ergodicity and the point spectrum of the Gaussian action
OpenQAlgorithms.Papers.Independent.Intakethird100.P38.gaussian_ergodicity_iff_point_spectrumLet , let have its full Schwartz Fréchet topology, and write . Let consist of the regular positive Borel measures for which for some nonnegative integer .
Let , let have the real characteristic functional of §1(d), and let be the full complex Koopman group of distribution shifts . Then
Frequency means for every . Ergodicity means that every measurable set invariant modulo null sets under all shifts has probability zero or one. The fixed space has complex dimension one, including when .
import Definitions.Def_QAlgorithms_Papers_Independent_Intakethird100_P38
noncomputable section
namespace QAlgorithms.Papers.Independent.Intakethird100.P38
open MeasureTheory Filter
open scoped Topology ComplexConjugate ENNReal
/-- Theorem 8.3, PDF pp.19--20: in this Gaussian distribution regime, ergodicity
is equivalent to absence of nonzero frequencies together with the one-dimensional
constant fixed space. This is not asserted for arbitrary probability actions. -/
theorem gaussian_ergodicity_iff_point_spectrum (n : ℕ) (hn : 0 < n)
(σ : Measure (SpectralDomain n)) (hσ : InC σ)
(μ : Measure (Omega n)) (hμ : CharacteristicLaw σ μ)
(U : KoopmanFamily μ) (hU : IsKoopman μ U) :
TrivialPointSpectrum μ U ↔ ShiftErgodic μ := by sorry
end QAlgorithms.Papers.Independent.Intakethird100.P38
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For every positive integer , let , and let be a regular measure satisfying for some natural number , possibly zero. Let be the continuous real-linear dual of the real-valued Schwartz space on , equipped with the sigma-algebra generated by evaluations. Suppose that is a probability measure on such that, for every real-valued Schwartz function , the extended nonnegative number
is finite and . Define . Let be a family of surjective complex-linear isometries of such that each is measurable and preserves , is the identity, for all , and almost everywhere for each . Then the following conjunction is equivalent to the stated property of measurable sets: for every and every nonzero ,
and, for every ,
Here the constant function has its usual class because is a probability measure; the underlying class construction returns zero for a function outside . The equivalent measurable-set property is that every measurable satisfying, for each , if and only if for -almost every , has or . The exceptional null set in this invariance hypothesis may depend on .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.