Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 16.4 — online linear regret bounds empirical Rademacher complexity

Proved
ModernOnlineLearning.ToX.theorem_16_4

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

online-learningp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-bookp2o-v1rademacher-complexityregret

Let Z⊆RdZ\subseteq\mathbb R^dZ⊆Rd be symmetric under negation, let z1,…,zT∈Zz_1,\ldots,z_T\in Zz1​,…,zT​∈Z, and let V⊆RdV\subseteq\mathbb R^dV⊆Rd be nonempty, bounded and closed. Suppose a deterministic online algorithm chooses xt∈Vx_t\in Vxt​∈V using only past gradients and has regret at most R(T)R(T)R(T) for every sequence of linear loss gradients in ZZZ. For the class Hlin={z↦⟨x,z⟩:x∈V}H_{\mathrm{lin}}=\{z\mapsto\langle x,z\rangle:x\in V\}Hlin​={z↦⟨x,z⟩:x∈V},

R^S(Hlin)=1TEε[sup⁡x∈V∑t=1Tεt⟨x,zt⟩]≤R(T)T.\widehat{\mathfrak R}_S(H_{\mathrm{lin}}) =\frac1T\mathbb E_\varepsilon\left[\sup_{x\in V}\sum_{t=1}^T\varepsilon_t\langle x,z_t\rangle\right] \le\frac{R(T)}T.RS​(Hlin​)=T1​Eε​[x∈Vsup​t=1∑T​εt​⟨x,zt​⟩]≤TR(T)​.

This reduction turns any uniform online linear regret bound into a bound on the complexity of the same linear class.

Formalization Note The algorithm is deterministic and nonanticipating; the book also permits independent internal randomization. The sample is indexed from one and T≥1T\ge1T≥1. The real supremum is taken over exactly VVV; boundedness and nonemptiness rule out default values on unbounded or empty sets.

Preamble
import Mathlib
import Definitions.Def_ModernOnlineLearning_ToX_Algorithms
Formal statement
namespace ModernOnlineLearning.ToX

/-- Theorem 16.4, p. 272: a deterministic non-anticipating online linear
algorithm with worst-case regret at most `R T` bounds the empirical
Rademacher complexity of the linear class indexed by `V`. -/
theorem theorem_16_4 {d T : ℕ} (hT : 1 ≤ T)
    (Z V : Set (Fin d → ℝ)) (z : ℕ → Fin d → ℝ)
    (hZsym : ∀ a ∈ Z, -a ∈ Z)
    (hz : ∀ t ∈ Finset.Icc 1 T, z t ∈ Z)
    (hVne : V.Nonempty) (hVbd : Bornology.IsBounded V) (hVclosed : IsClosed V)
    (R : ℕ → ℝ) (A : (ℕ → Fin d → ℝ) → ℕ → Fin d → ℝ)
    (hA : IsOnlineAlgorithm T Z V (R T) A) :
    empiricalRadLinear (T := T) V hT ⟨hVne, hVbd⟩ z ≤ R T / (T : ℝ) := by sorry

end ModernOnlineLearning.ToX
Source
Orabona, arXiv:1912.13213v10, Theorem 16.4, p. 272, displays (16.3)–(16.4)
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