Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Master decomposition (G.3)–(G.5) of the GIA-FTRL regret against an atom

Proved
IAFTRL.lemma_G_master

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

bandit-algorithmslinear-banditsonline-learning

Setting. A GIA-FTRL run of T≥1T\ge1T≥1 rounds against an adaptive environment, any vector θ\thetaθ, and any atom a∗a^*a∗ of the reference measure μ\muμ, with h∗=log⁡(1/μ({a∗}))h^*=\log(1/\mu(\{a^*\}))h∗=log(1/μ({a∗})).

Claim.

RegT(a∗)≤I1+I2+I3+I4.\mathrm{Reg}_T(a^*)\le I_1+I_2+I_3+I_4 .RegT​(a∗)≤I1​+I2​+I3​+I4​.

Here I1=β1h∗+∑t≥2(βt−βt−1)E[h∗−DKL(qt,μ)]I_1=\beta_1h^*+\sum_{t\ge2}(\beta_t-\beta_{t-1})\mathbb E[h^*-D_{\mathrm{KL}}(q_t,\mu)]I1​=β1​h∗+∑t≥2​(βt​−βt−1​)E[h∗−DKL​(qt​,μ)], I2=∑tE[Dt]I_2=\sum_t\mathbb E[D_t]I2​=∑t​E[Dt​], I3=∑tE⟨pt−qt,Δ⟩I_3=\sum_t\mathbb E\langle p_t-q_t,\Delta\rangleI3​=∑t​E⟨pt​−qt​,Δ⟩, and I4=∑tE⟨pt−qt,x⊤ctvec⟩I_4=\sum_t\mathbb E\langle p_t-q_t,x^\top c^{\mathrm{vec}}_t\rangleI4​=∑t​E⟨pt​−qt​,x⊤ctvec​⟩.

Proof idea.

  1. Unbiased loss differences give RegT(a∗)=E∑t⟨qt−δa∗,ℓ^t⟩+E∑t⟨pt−qt,x⊤θt⟩\mathrm{Reg}_T(a^*)=\mathbb E\sum_t\langle q_t-\delta_{a^*},\hat\ell_t\rangle+\mathbb E\sum_t\langle p_t-q_t,x^\top\theta_t\rangleRegT​(a∗)=E∑t​⟨qt​−δa∗​,ℓ^t​⟩+E∑t​⟨pt​−qt​,x⊤θt​⟩.
  2. The localized entropy identity (J.5) bounds the first sum by I1+I2I_1+I_2I1​+I2​.
  3. The second sum splits as θt=θ+ctvec\theta_t=\theta+c^{\mathrm{vec}}_tθt​=θ+ctvec​.
Preamble
import Mathlib
import Definitions.Def_IAFTRL_LinEnv
import Definitions.Def_IAFTRL_GIAFTRL
import Definitions.Def_IAFTRL_GTerms
Formal statement
namespace IAFTRL

open MeasureTheory

/-- **The master decomposition (G.3)–(G.5) with the localized entropy identity (J.5).** Zhao–Tan, *Best-of-Both-Worlds FTRL for Linear Bandits* (STOC submission version),
proof of Proposition G.1, pp.123–124: "`Reg_T(a*) = E[Σ_t ⟨q_t − δ_a*, ℓ̂_t⟩] + E[Σ_t ⟨p_t − q_t, Δ⟩] +
E[Σ_t ⟨p_t − q_t, xᵀc_t⟩] ≤ β₁h* + Σ_{t=2}^T (β_t − β_{t−1}) E[h* − D_KL(q_t, µ)] + Σ_t E[D_t] + …`"
(Lemma J.2, (J.5)).

For every execution of GIA-FTRL for `T ≥ 1` rounds, every vector `θ` and every atom `a*` of the
reference measure (`µ({a*}) > 0`, `h* = log(1/µ({a*}))`):
`Reg_T(a*) ≤ I₁ + I₂ + I₃ + I₄` with the sums of `IAFTRL_GTerms`. -/
theorem lemma_G_master {d : ℕ} (hd : 1 ≤ d) (𝒜 ℬ : Set (Fin d → ℝ)) (e : ℕ → Fin d → ℚ)
    (h𝒜 : IsCompact 𝒜) (hℬ : IsCompact ℬ) (hgen : IsGenerator 𝒜 ℬ)
    (hball : ∀ a ∈ 𝒜, a ⬝ᵥ a ≤ 1) (hspan : Submodule.span ℝ 𝒜 = ⊤) (hk : 1 ≤ kdim ℬ)
    {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω) [IsProbabilityMeasure P] (T : ℕ)
    (R : GRun d Ω) (hT : 1 ≤ T) (hR : IsGIAFTRLRun P 𝒜 ℬ e T R) (θ astar : Fin d → ℝ)
    (hatom : 0 < refMu ℬ e {astar}) :
    gRegretVs P T R astar ≤ penI1 P ℬ e T R astar + stabI2 P ℬ e T R +
      exploI3 P ℬ e T R θ astar + corrI4 P ℬ e T R θ := 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 G.1.1, (G.3)–(G.5), and Lemma J.2 (J.5)

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