Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

From gap-dependent pull bounds to minimax regret

Proved
BanditAlgorithm.finite_gap_pull_bound_to_minimax

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

banditsinequalitiesminimaxregret

Let k≥1k\ge1k≥1 and n≥2n\ge2n≥2. For each arm iii, let its gap satisfy 0≤Δi≤10\le\Delta_i\le10≤Δi​≤1, and let Ti≥0T_i\ge0Ti​≥0 be an expected pull count with ∑iTi=n\sum_iT_i=n∑i​Ti​=n. Suppose that every positive-gap arm satisfies the logarithmic instance-dependent estimate

Ti≤C(1+log⁡nΔi2),C≥0.T_i\le C\left(1+\frac{\log n}{\Delta_i^2}\right), \qquad C\ge0.Ti​≤C(1+Δi2​logn​),C≥0.

Then the corresponding regret obeys the distribution-free bound

∑iΔiTi≤(1+3C)knlog⁡n.\sum_i\Delta_iT_i \le(1+3C)\sqrt{kn\log n}.i∑​Δi​Ti​≤(1+3C)knlogn​.

This is the standard small-gap/large-gap conversion: split at klog⁡(n)/n\sqrt{k\log(n)/n}klog(n)/n​, control small gaps using the total pull count, and control large gaps using the instance-dependent estimate.

Preamble
import Mathlib.Analysis.SpecialFunctions.Log.Basic
import Mathlib.Analysis.SpecialFunctions.Pow.Real
import Mathlib.Analysis.Complex.ExponentialBounds
import Mathlib.Algebra.Order.BigOperators.Group.Finset

open scoped BigOperators
Formal statement
theorem BanditAlgorithm.finite_gap_pull_bound_to_minimax {k n : ℕ} [NeZero k]
    (hn : 2 ≤ n) (Δ T : Fin k → ℝ) (C : ℝ)
    (hΔ : ∀ i, Δ i ∈ Set.Icc (0 : ℝ) 1)
    (hT : ∀ i, 0 ≤ T i) (hsum : ∑ i, T i = n)
    (hC : 0 ≤ C)
    (hpull : ∀ i, 0 < Δ i →
      T i ≤ C * (1 + Real.log n / Δ i ^ 2)) :
    ∑ i, Δ i * T i ≤
      (1 + 3 * C) * Real.sqrt (k * n * Real.log n) := by
  sorry
Source
Lattimore and Szepesvari, Bandit Algorithms (CUP 2020), Theorem 36.3, printed p.465, distribution-free conclusion; standard small-gap/large-gap conversion used throughout Chapters 7-9.

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