Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 5.4, p. 21 — uniform empirical small-ball estimate on the L₂ unit sphere

Proved
LearnNoConc.ERM.theorem_5_4

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

empirical-processesp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1small-ball-methodstatistical-learning

Let (Ω,μ)(\Omega,\mu)(Ω,μ) be a probability space and let HHH be a class of measurable functions on the unit sphere of L2(μ)L_2(\mu)L2​(μ) (∥h∥L2=1\|h\|_{L_2}=1∥h∥L2​​=1 for every h∈Hh\in Hh∈H). Let X1,…,XNX_1,\dots,X_NX1​,…,XN​ (N≥1N\ge1N≥1) be i.i.d. with law μ\muμ and (εi)(\varepsilon_i)(εi​) independent random signs. Suppose that τ>0\tau>0τ>0 satisfies QH(2τ)>0Q_H(2\tau)>0QH​(2τ)>0, where QH(u)=inf⁡h∈HPr⁡(∣h∣≥u∥h∥L2)Q_H(u)=\inf_{h\in H}\Pr(|h|\ge u\|h\|_{L_2})QH​(u)=infh∈H​Pr(∣h∣≥u∥h∥L2​​), and that

Esup⁡h∈H∣1N∑i=1Nεih(Xi)∣≤τQH(2τ)16.\mathbb E\sup_{h\in H}\Bigl|\frac1N\sum_{i=1}^N\varepsilon_i h(X_i)\Bigr|\le\frac{\tau Q_H(2\tau)}{16}.Eh∈Hsup​​N1​i=1∑N​εi​h(Xi​)​≤16τQH​(2τ)​.

Then, with probability at least 1−2exp⁡(−NQH2(2τ)/2)1-2\exp(-NQ_H^2(2\tau)/2)1−2exp(−NQH2​(2τ)/2),

inf⁡h∈H∣{i:∣h(Xi)∣≥τ}∣≥N QH(2τ)4.\inf_{h\in H}\bigl|\{i:|h(X_i)|\ge\tau\}\bigr|\ge N\,\frac{Q_H(2\tau)}{4}.h∈Hinf​​{i:∣h(Xi​)∣≥τ}​≥N4QH​(2τ)​.

This is the uniform lower estimate that drives the proof of Theorem 3.1: under a small-ball condition, every function of the class is large on a proportional number of sample points, without any concentration assumption.

Formalization Note The conclusion is stated as: the (outer) probability that some h∈Hh\in Hh∈H has fewer than NQH(2τ)/4NQ_H(2\tau)/4NQH​(2τ)/4 indices with ∣h(Xi)∣≥τ|h(X_i)|\ge\tau∣h(Xi​)∣≥τ is at most 2exp⁡(−NQH2(2τ)/2)2\exp(-NQ_H^2(2\tau)/2)2exp(−NQH2​(2τ)/2). The expectation of the supremum is taken in [0,∞][0,\infty][0,∞]. Measurability of every hhh, membership in L2(μ)L_2(\mu)L2​(μ), and pointwise separability of HHH (a countable subclass approximating every element pointwise and in L2L_2L2​) are added; the last stands in for the measurability of suprema that the paper takes for granted.

Preamble
import Mathlib
import Definitions.Def_LearnNoConc_ERM_Setting
Formal statement
namespace LearnNoConc.ERM

open MeasureTheory
open scoped ENNReal

/-- Theorem 5.4, p. 21: the uniform empirical small-ball estimate on the `L₂(μ)` unit sphere. -/
theorem theorem_5_4 {Ω : Type*} [MeasurableSpace Ω] (μ : Measure Ω) [IsProbabilityMeasure μ]
    (H : Set (Ω → ℝ)) (hHm : ∀ h ∈ H, Measurable h) (hH2 : ∀ h ∈ H, MemLp h 2 μ)
    (hS : ∀ h ∈ H, eLpNorm h 2 μ = 1) (hHsep : PointwiseSeparable μ H)
    (N : ℕ) (hN : 0 < N) (τ : ℝ) (hτ : 0 < τ) (hQ : 0 < Q μ H (2 * τ))
    (hE : radSup μ H N ≤ ENNReal.ofReal (τ * Q μ H (2 * τ) / 16)) :
    (Measure.pi fun _ : Fin N => μ)
      {x | ∃ h ∈ H, ((Finset.univ.filter fun i => τ ≤ |h (x i)|).card : ℝ)
          < N * Q μ H (2 * τ) / 4}
      ≤ ENNReal.ofReal (2 * Real.exp (-(N * Q μ H (2 * τ) ^ 2 / 2))) := by sorry

end LearnNoConc.ERM
Source
Mendelson, Learning without Concentration, arXiv:1401.0304v2, Theorem 5.4, p. 21
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