Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

One-round Thompson-sampling information-ratio bound

Proved
BanditAlgorithm.bayesian_ts_round_regret_sq_le_mutualInformation_increment

by Harry_Xu · Aug 1, 2026 · Mathlib 0df444a (Lean v4.33.1)

banditsinformation-ratiothompson-sampling

Let

Δt=E ⁣[Xt,A∗−Xt,At]\Delta_t=\mathbb E\!\left[X_{t,A^*}-X_{t,A_t}\right]Δt​=E[Xt,A∗​−Xt,At​​]

be the one-round Bayesian regret of posterior-sampling Thompson sampling, and let It=I(Ht;A∗)I_t=I(H_t;A^*)It​=I(Ht​;A∗) be the mutual information between the history before round ttt and the optimal action. If rewards lie in [0,1][0,1][0,1], then

Δt2≤k2(It+1−It).\Delta_t^2\le \frac{k}{2}\bigl(I_{t+1}-I_t\bigr).Δt2​≤2k​(It+1​−It​).

This is the one-step information-ratio estimate: the squared expected regret is controlled by the information about A∗A^*A∗ revealed by the next action–reward observation.

Preamble
import Definitions.Def_BayesianHistoryMutualInformation

open MeasureTheory ProbabilityTheory
Formal statement
theorem BanditAlgorithm.bayesian_ts_round_regret_sq_le_mutualInformation_increment
    {k n : ℕ} [NeZero k]
    (Q : Measure (Fin n → Fin k → ℝ)) [IsProbabilityMeasure Q]
    (hQ : ∀ᵐ X ∂Q, ∀ (s : Fin n) (a : Fin k), X s a ∈ Set.Icc (0 : ℝ) 1)
    {pi : BanditPolicy k} (hpi : IsBayesianTSPolicy Q pi) (t : Fin n) :
    (∫ p, (p.1 t (bayesianOptimalAction p.1) - p.1 t ((p.2 t).1))
      ∂bayesianAdversarialMeasure Q pi n le_rfl) ^ 2 ≤
      ((k : ℝ) / 2) *
        (bayesianHistoryMutualInformation Q pi (t.1 + 1) (Nat.succ_le_iff.mpr t.2) -
          bayesianHistoryMutualInformation Q pi t.1 (Nat.le_of_lt t.2)) := by
  sorry
Source
Lattimore and Szepesvári, Bandit Algorithms (Cambridge UP, 2020), https://tor-lattimore.com/downloads/book/book.pdf, printed pp. 470–471 (PDF pp. 479–480), Lemma 36.7 and the Bayes-rule equality at the end of its proof.

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