Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Stopped relative entropy is approximated by bounded stopping times

Proved
BanditAlgorithm.klDiv_stopped_le_iSup_truncations

by Harry_Xu · Jul 30, 2026 · Mathlib c5ea003 (Lean v4.30.0)

banditsinformation-theorymeasure-theorystopping-times

Run the same adaptive bandit policy in two environments, producing trajectory laws PPP and QQQ, and let τ\tauτ be a stopping time with finite expectation under PPP. Write Fτ\mathcal F_\tauFτ​ for the stopped sigma-algebra and τn=min⁡{τ,n}\tau_n=\min\{\tau,n\}τn​=min{τ,n}. Then the relative entropy of the stopped experiment is bounded by the supremum of the relative entropies of its bounded truncations:

D ⁣(P∣Fτ ∥ Q∣Fτ)≤sup⁡n≥0D ⁣(P∣Fτn ∥ Q∣Fτn).D\!\left(P|_{\mathcal F_\tau}\,\middle\Vert\,Q|_{\mathcal F_\tau}\right) \le \sup_{n\ge0}D\!\left(P|_{\mathcal F_{\tau_n}}\,\middle\Vert\,Q|_{\mathcal F_{\tau_n}}\right).D(P∣Fτ​​∥Q∣Fτ​​)≤n≥0sup​D(P∣Fτn​​​​Q∣Fτn​​​).

This is the measure-theoretic truncation principle needed to pass finite-horizon likelihood calculations to an integrable, possibly unbounded stopping time. It is useful beyond best-arm identification whenever an adaptive experiment is observed at a random horizon.

Formalization Note The trajectory measures are restricted with Measure.trim to Mathlib's stopped measurable spaces. The supremum is represented as an ENNReal indexed supremum.

Preamble
import Definitions.Def_BanditTrajectory

open MeasureTheory ProbabilityTheory InformationTheory ENNReal
Formal statement
theorem BanditAlgorithm.klDiv_stopped_le_iSup_truncations
    {k : ℕ} (ν ν' : StochasticBandit k) (π : BanditPolicy k)
    (τ : (ℕ → Fin k × ℝ) → ℕ∞) (hτ : IsBanditStoppingTime τ)
    (hfinite : ∫⁻ ω, (τ ω : ℝ≥0∞) ∂banditTrajMeasure ν π < ⊤) :
    @klDiv (ℕ → Fin k × ℝ) hτ.measurableSpace
        ((banditTrajMeasure ν π).trim hτ.measurableSpace_le)
        ((banditTrajMeasure ν' π).trim hτ.measurableSpace_le) ≤
      ⨆ n : ℕ,
        @klDiv (ℕ → Fin k × ℝ) (hτ.min_const n).measurableSpace
          ((banditTrajMeasure ν π).trim (hτ.min_const n).measurableSpace_le)
          ((banditTrajMeasure ν' π).trim (hτ.min_const n).measurableSpace_le) := by
  sorry
Source
Lattimore and Szepesvári, Bandit Algorithms (CUP 2020), Exercise 15.7, printed p. 211, truncation step extending Lemma 15.1 to stopping times; stopped sigma-algebra formulation as used in Chapter 33, Section 33.2.

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