Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Section 2, Eq. (14) — under Rˉ\bar RRˉ, Sn/n→max⁡(α,β)S_n/n \to \max(\alpha,\beta)Sn​/n→max(α,β) with probability 1

Proved
RobbinsSeqDesign.ForcedSampling.average_reward_tendsto_max_ae

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

bandit-algorithmsp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1sequential-designstrong-law

Let AAA and BBB be two populations whose expectations α\alphaα and β\betaβ exist (are finite), let 1=a1<a2<⋯1 = a_1 < a_2 < \cdots1=a1​<a2​<⋯ and 2=b1<b2<⋯2 = b_1 < b_2 < \cdots2=b1​<b2​<⋯ be disjoint sequences of density 000, and sample according to Robbins' rule Rˉ\bar RRˉ: forced draws from AAA at the times aja_jaj​ and from BBB at the times bjb_jbj​, and otherwise a draw from the population whose running mean is larger (ties to BBB). With Sn=x1+⋯+xnS_n = x_1 + \cdots + x_nSn​=x1​+⋯+xn​, with probability 1,

lim⁡n→∞Snn=max⁡(α,β).\lim_{n\to\infty} \frac{S_n}{n} = \max(\alpha, \beta).n→∞lim​nSn​​=max(α,β).

This is the almost-sure form of the statement that Rˉ\bar RRˉ asymptotically does as well, per draw, as an experimenter who knows which population is better. The paper attributes it to the strong law of large numbers.

Formalization Note. "The expectations exist" is the hypothesis that both reward laws are integrable; no further condition (variance, boundedness) is assumed, and α=β\alpha=\betaα=β is allowed. "With probability 1" is an almost-everywhere statement under the trajectory law banditTrajMeasure ν (rule σ).

Preamble
import Mathlib
import Definitions.Def_RobbinsSeqDesign_ForcedSampling_Rule
import Definitions.Def_RobbinsSeqDesign_ForcedSampling_Loss
Formal statement
namespace RobbinsSeqDesign.ForcedSampling

open MeasureTheory ProbabilityTheory Filter Topology BanditAlgorithm

/-- Robbins (1952), Section 2, Eq. (14), p. 532: for any two populations A, B whose
expectations `α`, `β` exist (are finite) and any pair of sequences (13), under the rule `R̄`,
with probability 1, `S_n / n → max(α, β)`. -/
theorem average_reward_tendsto_max_ae (ν : StochasticBandit 2)
    (hint : ∀ i, Integrable id (ν.P i)) (σ : ForcedSchedule) :
    ∀ᵐ ω ∂(banditTrajMeasure ν (rule σ)),
      Tendsto (fun n : ℕ => partialSum n ω / (n : ℝ)) atTop
        (𝓝 (max (banditArmMean ν 0) (banditArmMean ν 1))) := by sorry

end RobbinsSeqDesign.ForcedSampling
Source
Robbins, Some aspects of the sequential design of experiments, Bull. Amer. Math. Soc. 58 (1952), p. 532, Section 2, Eq. (14)
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