Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Equation (2.5), p. 8 — empirical gap lower tail

Proved
BanditCovariates.SE.bound_2_5

by mikedeng1 · Oct 8, 2026 · Mathlib 0df444a (Lean v4.33.1)

multi-armed-banditp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1probabilitysuccessive-elimination

In the sorted K+1K+1K+1 arm model, let iii be a suboptimal arm with gap Δi=f∗−fi\Delta_i=f^*-f_iΔi​=f∗−fi​. Let Δ^i(τ)\widehat\Delta_i(\tau)Δi​(τ) be the difference of the empirical means of the best arm and arm iii after τ≥1\tau\ge1τ≥1 samples each. If γU(τ,T)≤Δi\gamma U(\tau,T)\le\Delta_iγU(τ,T)≤Δi​, then

P{Δ^i(τ)<γU(τ,T)}≤exp⁡ ⁣(−τ(Δi−γU(τ,T))22).P\{\widehat\Delta_i(\tau)<\gamma U(\tau,T)\}\le\exp\!\left(-\frac{\tau(\Delta_i-\gamma U(\tau,T))^2}{2}\right).P{Δi​(τ)<γU(τ,T)}≤exp(−2τ(Δi​−γU(τ,T))2​).

This lower-tail estimate controls failure to eliminate a sufficiently separated arm.

Formalization Note The printed phrase “for every τ≥1\tau\ge1τ≥1” needs the displayed threshold condition: a negative deviation size cannot be inserted into the stated Hoeffding bound. The proof uses the estimate where this condition holds. Rewards are bounded almost surely, and the published reward-stack model supplies independence.

Preamble
import Mathlib
import Definitions.Def_BanditCovariates_SE_Setting
Formal statement
namespace BanditCovariates.SE

open MeasureTheory ProbabilityTheory

/-- Equation (2.5), p. 8, in the range where the displayed Hoeffding estimate applies. -/
theorem bound_2_5 {Ω : Type*} {m : MeasurableSpace Ω} (P : Measure Ω)
    [IsProbabilityMeasure P] (K : ℕ) (hK : 1 ≤ K)
    (Y : Fin (K + 1) → ℕ → Ω → ℝ) (f : Fin (K + 1) → ℝ)
    (hmodel : RegretBandits.Stochastic.IsStochasticBandit P Y f)
    (hbounded : ∀ i k, ∀ᵐ ω ∂P, Y i k ω ∈ Set.Icc (0 : ℝ) 1)
    (hordered : Monotone f)
    (hbest : ∀ i, i ≠ Fin.last K → f i < f (Fin.last K))
    (i : Fin (K + 1)) (hi : i ≠ Fin.last K)
    (T τ : ℕ) (hT : 1 ≤ T) (hτ : 1 ≤ τ) (γ : ℝ) (hγ : 1 ≤ γ)
    (hthreshold : γ * U τ T ≤ gap f i) :
    P.real {ω | estimatedGap Y i τ ω < γ * U τ T} ≤
      Real.exp (-(τ : ℝ) * (gap f i - γ * U τ T) ^ 2 / 2) := by sorry

end BanditCovariates.SE
Source
Perchet, Rigollet, The multi-armed bandit problem with covariates, arXiv:1110.6084v3, p. 8, (2.5)
Human review
  • Endorsed by Shuze Chen · Oct 8, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 8, 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