Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Asymptotic pull count of a suboptimal arm under Gaussian TS

Proved
BanditAlgorithm.gaussian_ts_suboptimal_pull_count_asymptotic

by Harry_Xu · Jul 30, 2026 · Mathlib 0df444a (Lean v4.33.1)

banditsgaussianprobabilitythompson-sampling

Let Thompson sampling use independent Gaussian posterior perturbations of variance 1/Ti(t)1/T_i(t)1/Ti​(t) in a unit-variance Gaussian kkk-armed bandit. For any suboptimal arm iii with gap Δi>0\Delta_i>0Δi​>0, its expected number of pulls satisfies

lim⁡n→∞E[Ti(n)]log⁡n=2Δi2.\lim_{n\to\infty}\frac{\mathbb E[T_i(n)]}{\log n}=\frac{2}{\Delta_i^2}.n→∞lim​lognE[Ti​(n)]​=Δi2​2​.

This is the armwise probabilistic core of the instance-dependent asymptotic regret theorem. Summing it against the gaps via the regret decomposition yields the Lai--Robbins-optimal coefficient ∑i:Δi>02/Δi\sum_{i:\Delta_i>0}2/\Delta_i∑i:Δi​>0​2/Δi​.

Preamble
import Definitions.Def_ThompsonSampling
import Definitions.Def_GaussianBandit

open MeasureTheory ProbabilityTheory Filter
Formal statement
theorem BanditAlgorithm.gaussian_ts_suboptimal_pull_count_asymptotic
    {k : ℕ} [NeZero k] (μvec : Fin k → ℝ) (π : BanditPolicy k)
    (hπ : IsGaussianTSPolicy π) (i : Fin k)
    (hi : 0 < banditGap (gaussianBandit μvec) i) :
    Tendsto
      (fun n : ℕ ↦
        (∫ h, (armPullCount i h : ℝ)
          ∂banditMeasure (gaussianBandit μvec) π n) / Real.log n)
      atTop
      (nhds (2 / banditGap (gaussianBandit μvec) i ^ 2)) := by
  sorry
Source
Lattimore and Szepesvari, Bandit Algorithms (CUP 2020), Theorem 36.3 and its proof, printed pp. 464-468, using Theorem 36.2.

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