Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

One-step Thompson-sampling diagonal information-ratio bound

Proved
BanditAlgorithm.thompson_sampling_one_step_information_ratio_diagonal

by Harry_Xu · Aug 1, 2026 · Mathlib c5ea003 (Lean v4.30.0)

banditsinformation-ratiothompson-sampling

Fix a posterior weight vector ppp on kkk actions. For each action aaa, let PaP_aPa​ be the conditional observation law on the event that aaa is optimal, let MMM be the posterior-predictive observation law, and let ra∈[0,1]r_a\in[0,1]ra​∈[0,1] be the reward observable. If every D(Pa ∥ M)D(P_a\,\|\,M)D(Pa​∥M) is finite, then

[∑apa(EPara−EMra)]2≤k2∑apa2D(Pa ∥ M).\left[\sum_a p_a\left(\mathbb E_{P_a}r_a-\mathbb E_M r_a\right)\right]^2\le \frac{k}{2}\sum_a p_a^2D(P_a\,\|\,M).[a∑​pa​(EPa​​ra​−EM​ra​)]2≤2k​a∑​pa2​D(Pa​∥M).

The right side is the diagonal contribution to the information gained by posterior sampling.

Preamble
import Mathlib.InformationTheory.KullbackLeibler.Basic

open MeasureTheory InformationTheory
open scoped ENNReal BigOperators
Formal statement
theorem BanditAlgorithm.thompson_sampling_one_step_information_ratio_diagonal
    {k : ℕ} {Omega : Type} {mOmega : MeasurableSpace Omega}
    (p : Fin k → ℝ) (P : Fin k → Measure Omega)
    [∀ a, IsProbabilityMeasure (P a)]
    (M : Measure Omega) [IsProbabilityMeasure M]
    (reward : Fin k → Omega → ℝ)
    (hreward : ∀ a, Measurable (reward a))
    (hreward0 : ∀ a x, 0 ≤ reward a x)
    (hreward1 : ∀ a x, reward a x ≤ 1)
    (hfinite : ∀ a, klDiv (P a) M ≠ ∞) :
    (∑ a, p a * ((∫ x, reward a x ∂(P a)) - ∫ x, reward a x ∂M)) ^ 2 ≤
      ((k : ℝ) / 2) * ∑ a, p a ^ 2 * (klDiv (P a) M).toReal := by
  sorry
Source
Lattimore and Szepesvári, Bandit Algorithms (Cambridge UP, 2020), https://tor-lattimore.com/downloads/book/book.pdf, printed p. 470 (PDF p. 479), Lemma 36.7 and its proof: Pinsker followed by Cauchy–Schwarz, before the final Bayes-rule identification.

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