Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 2 — convergence to uniformity from every starting point

Proved
SmithHitAndRun.SymMixing.theorem_2_strongly_mixing

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

convergencemarkov-chainsmcmcp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

Let VVV be finite nonzero content on SSS and PPP a Markov kernel with a jointly measurable symmetric density f(y∣x)f(y\mid x)f(y∣x) relative to VVV. Suppose f(y∣x)>0f(y\mid x)>0f(y∣x)>0 for every pair x,y∈Sx,y\in Sx,y∈S. Write λ=V/V(S)\lambda=V/V(S)λ=V/V(S) and Pm(x,A)P^m(x,A)Pm(x,A) for the mmm-step transition probability. Then for every starting point x∈Sx\in Sx∈S and every measurable A⊆SA\subseteq SA⊆S,

lim⁡m→∞Pm(x,A)=λ(A).\lim_{m\to\infty}P^m(x,A)=\lambda(A).m→∞lim​Pm(x,A)=λ(A).

This is the paper's meaning of strongly mixing: setwise convergence to the uniform law from each point. It supplies the qualitative convergence target behind the Random Directions Algorithm. Formalization Note The abstract measurable space represents the whole region SSS and VVV its content; symmetry of fff encodes the consequence Smith asserts for symmetric mixing algorithms satisfying Assumption (a). The statement asks for every xxx, and does not assume stationarity, uniqueness, or a convergence bound.

Preamble
import Mathlib.Probability.Kernel.Invariance
import Definitions.Def_MarkovIterKernel
import Definitions.Def_SmithHitAndRun_SymMixing_uniformLaw
import Definitions.Def_SmithHitAndRun_SymMixing_IsSymmetricMixingKernel

open MeasureTheory ProbabilityTheory Filter
Formal statement
namespace SmithHitAndRun.SymMixing

/-- Smith (1984), Theorem 2, p. 1301: setwise convergence from every start. -/
theorem theorem_2_strongly_mixing {α : Type*} [MeasurableSpace α]
    (V : Measure α) [IsFiniteMeasure V] (hV : V ≠ 0)
    (P : Kernel α α) [IsMarkovKernel P] (f : α → α → ENNReal)
    (h : IsSymmetricMixingKernel V P f) (hpos : ∀ x y, 0 < f y x) :
    ∀ x : α, ∀ A : Set α, MeasurableSet A →
      Tendsto (fun m : ℕ => MarkovChainCLT.iterKernel P m x A)
        atTop (nhds (uniformLaw V A)) := by sorry

end SmithHitAndRun.SymMixing
Source
Smith, Efficient Monte Carlo Procedures for Generating Points Uniformly Distributed over Bounded Regions, Oper. Res. 32(6) (1984), p. 1301, Theorem 2
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me