Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma J.10 — exact compact-generator transfer

Proved
IAFTRL.lemma_J_10

by Canzhe Zhao · Nov 17, 2026 · Mathlib 0df444a (Lean v4.33.1)

bandit-algorithmslinear-banditsonline-learning

Let B⊆A⊆Conv(B)\mathcal B\subseteq\mathcal A\subseteq\mathrm{Conv}(\mathcal B)B⊆A⊆Conv(B) be compact subsets of Rd\mathbb R^dRd.

  1. For every z∈Rdz\in\mathbb R^dz∈Rd,
sup⁡a∈A∣a⊤z∣=sup⁡b∈B∣b⊤z∣.\sup_{a\in\mathcal A}|a^\top z|=\sup_{b\in\mathcal B}|b^\top z|.a∈Asup​∣a⊤z∣=b∈Bsup​∣b⊤z∣.
  1. If θ\thetaθ has a unique optimal action a∗∈Aa^*\in\mathcal Aa∗∈A, then cctr∗(A,θ)=cctr∗(B,θ)c^*_{\mathrm{ctr}}(\mathcal A,\theta)=c^*_{\mathrm{ctr}}(\mathcal B,\theta)cctr∗​(A,θ)=cctr∗​(B,θ).
  2. Consider any run whose physical actions lie in B\mathcal BB, with measurable actions and loss vectors satisfying ∥θt∥A≤1\|\theta_t\|_{\mathcal A}\le1∥θt​∥A​≤1. Then
sup⁡a∈ARegT(a)=sup⁡b∈BRegT(b).\sup_{a\in\mathcal A}\mathrm{Reg}_T(a)=\sup_{b\in\mathcal B}\mathrm{Reg}_T(b).a∈Asup​RegT​(a)=b∈Bsup​RegT​(b).

So running on a generator loses nothing: the regret, the corruption norm and the information-acquisition complexity are the same whether they are measured on A\mathcal AA or on B\mathcal BB.

Preamble
import Mathlib
import Definitions.Def_IAFTRL_LinEnv
import Definitions.Def_IAFTRL_GIAFTRL
Formal statement
namespace IAFTRL

open MeasureTheory

/-- **Lemma J.10 (Exact compact-generator transfer).** Zhao–Tan, *Best-of-Both-Worlds FTRL for
Linear Bandits* (STOC submission version), Lemma J.10, p.209: "If known compact sets obey
`ℬ ⊆ 𝒜 ⊆ Conv(ℬ)`, then a strategy playing only `ℬ` satisfies
`sup_{a∈𝒜} Reg_T(a) = sup_{b∈ℬ} Reg_T(b)`, `sup_{a∈𝒜} |aᵀz| = sup_{b∈ℬ} |bᵀz|`.
For a unique optimal action `a*` under `θ`, `c*_ctr(𝒜, θ) = c*_ctr(ℬ, θ)`."

1. `‖z‖_𝒜 = ‖z‖_ℬ` for every `z`;
2. for a unique optimal action `a* ∈ 𝒜` under `θ`, `c*_ctr(𝒜, θ) = c*_ctr(ℬ, θ)`;
3. for every run whose physical actions lie in `ℬ` (measurable actions and loss vectors with
   `‖θ_t‖_𝒜 ≤ 1`), the suprema of the fixed-comparator regret over `𝒜` and over `ℬ` coincide. -/
theorem lemma_J_10 {d : ℕ} (𝒜 ℬ : Set (Fin d → ℝ)) (h𝒜 : IsCompact 𝒜) (hℬ : IsCompact ℬ)
    (hgen : IsGenerator 𝒜 ℬ) :
    (∀ z : Fin d → ℝ, setNorm 𝒜 z = setNorm ℬ z) ∧
    (∀ θ astar : Fin d → ℝ, astar ∈ 𝒜 → (∀ a ∈ 𝒜, a ≠ astar → astar ⬝ᵥ θ < a ⬝ᵥ θ) →
      cstarCtr 𝒜 θ astar = cstarCtr ℬ θ astar) ∧
    (∀ {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω) [IsProbabilityMeasure P] (T : ℕ)
      (R : GRun d Ω), (∀ t < T, Measurable (R.A t) ∧ Measurable (R.θs t) ∧
        (∀ ω, R.A t ω ∈ ℬ) ∧ ∀ ω, setNorm 𝒜 (R.θs t ω) ≤ 1) →
      sSup (gRegretVs P T R '' 𝒜) = sSup (gRegretVs P T R '' ℬ)) := by sorry

end IAFTRL
Source
Zhao–Tan, Best-of-Both-Worlds FTRL for Linear Bandits: The Interplay of Information Acquisition, Corruption, and Geometry (STOC submission version; arXiv:2611.01299), Appendix J.2, Lemma J.10, p. 209

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