Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 13.9, p. 215 — coordinate-wise KT parameter-free regret

Proved
ModernOnlineLearning.ParameterFree.theorem_13_9

by mikedeng1 · Oct 9, 2026 · Mathlib 0df444a (Lean v4.33.1)

coin-bettingonline-learningp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-bookp2o-v1parameter-free-regret

Let d≥1d\ge1d≥1, ε>0\varepsilon>0ε>0, L∞>0L_\infty>0L∞​>0, and T≥1T\ge1T≥1. Run Algorithm 13.3 on subdifferentiable losses ℓt:Rd→R\ell_t:\mathbb R^d\to\mathbb Rℓt​:Rd→R, choosing a full-space subgradient gtg_tgt​ at each prediction xtx_txt​ with ∥gt∥∞≤L∞\|g_t\|_\infty\le L_\infty∥gt​∥∞​≤L∞​. For every u∈Rdu\in\mathbb R^du∈Rd, both inequalities hold:

∑t=1T(ℓt(xt)−ℓt(u))≤L∞∑i=1d∣ui∣2Tln⁡(1+e∣ui∣Tε)+dεL∞≤∥u∥1L∞2Tln⁡(1+e∥u∥∞Tε)+dεL∞.\sum_{t=1}^T\bigl(\ell_t(x_t)-\ell_t(u)\bigr)\le L_\infty\sum_{i=1}^d |u_i|\sqrt{2T\ln\left(1+\frac{e|u_i|T}{\varepsilon}\right)}+d\varepsilon L_\infty\le\|u\|_1L_\infty\sqrt{2T\ln\left(1+\frac{e\|u\|_\infty T}{\varepsilon}\right)}+d\varepsilon L_\infty.t=1∑T​(ℓt​(xt​)−ℓt​(u))≤L∞​i=1∑d​∣ui​∣2Tln(1+εe∣ui​∣T​)​+dεL∞​≤∥u∥1​L∞​2Tln(1+εe∥u∥∞​T​)​+dεL∞​.

The result gives one parameter-free regret bound for each coordinate and the resulting bound in ℓ1\ell_1ℓ1​ and ℓ∞\ell_\inftyℓ∞​ norms.

Formalization Note Vectors are functions on Fin d; their function-space norm is the sup norm. The run predicate enforces Algorithm 13.3's exact update and bounds each selected subgradient coordinate. Positivity of ε\varepsilonε, L∞L_\inftyL∞​, and TTT is explicit.

Preamble
import Mathlib
import Definitions.Def_ModernOnlineLearning_ParameterFree_Defs
Formal statement
namespace ModernOnlineLearning.ParameterFree

/-- Theorem 13.9, p. 215: both coordinate-wise and norm-form regret bounds. -/
theorem theorem_13_9 (d : ℕ) (hd : 0 < d) (ε L : ℝ)
    (hε : 0 < ε) (hL : 0 < L) (T : ℕ) (hT : 1 ≤ T)
    (ℓ : ℕ → (Fin d → ℝ) → ℝ) (g x : ℕ → Fin d → ℝ)
    (hrun : IsCoordKTRun ε L T ℓ g x) :
    ∀ u : Fin d → ℝ,
      (∑ t ∈ Finset.Icc 1 T, (ℓ t (x t) - ℓ t u)) ≤
        L * (∑ i : Fin d, |u i| * Real.sqrt
          (2 * (T : ℝ) * Real.log
            (1 + Real.exp 1 * |u i| * (T : ℝ) / ε))) + (d : ℝ) * ε * L ∧
      L * (∑ i : Fin d, |u i| * Real.sqrt
        (2 * (T : ℝ) * Real.log
          (1 + Real.exp 1 * |u i| * (T : ℝ) / ε))) + (d : ℝ) * ε * L ≤
        (∑ i : Fin d, |u i|) * L * Real.sqrt
          (2 * (T : ℝ) * Real.log
            (1 + Real.exp 1 * ‖u‖ * (T : ℝ) / ε)) + (d : ℝ) * ε * L := by sorry

end ModernOnlineLearning.ParameterFree
Source
Orabona, arXiv:1912.13213v10, Theorem 13.9, p. 215
Human review
  • Endorsed by Shuze Chen · Oct 9, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 9, 2026

    Confirmed by the mission captain (proposal self-audit).

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