Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Appendix D, p. 18 — adversarial hinge objective in closed form

Proved
RobustOdds.SVM.advMarginLoss_eq

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

adversarial-robustnessclassificationgaussian-modelp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

For ε≥0\varepsilon\ge0ε≥0, let Ladv(w)L_{\mathrm{adv}}(w)Ladv​(w) be the expectation of the largest hinge loss over ℓ∞\ell_\inftyℓ∞​ perturbations of radius ε\varepsilonε, with the maximum taken separately for each input and label. Then

Ladv(w)=ED ⁣[max⁡{0,1−yw⊤x+ε∑i∣wi∣}].L_{\mathrm{adv}}(w)=\mathbb E_D\!\left[\max\{0,1-yw^\top x+\varepsilon\sum_i|w_i|\}\right].Ladv​(w)=ED​[max{0,1−yw⊤x+εi∑​∣wi​∣}].

Thus the adversary contributes the ℓ1\ell_1ℓ1​ norm of the weights. Adding the same quadratic regularizer to both sides gives the objective identity used in the proof of Lemma D.5.

Formalization Note Equation (2) puts the maximum inside expectation. The display in the proof of Lemma D.5 prints it outside; that placement does not yield this identity.

Preamble
import Mathlib
import Definitions.Def_RobustOdds_SVM_Setting
Formal statement
namespace RobustOdds.SVM

open MeasureTheory

theorem advMarginLoss_eq (d : ℕ) (p η ε : ℝ) (hε : 0 ≤ ε)
    (w : Fin (d + 1) → ℝ) :
    advMarginLoss d p η ε w =
      (1 / 2) * ∫ x, max 0 (1 - ∑ i, w i * x i + ε * ∑ i, |w i|)
        ∂(RobustOdds.Tradeoff.condLaw d p η 1) +
      (1 / 2) * ∫ x, max 0 (1 + ∑ i, w i * x i + ε * ∑ i, |w i|)
        ∂(RobustOdds.Tradeoff.condLaw d p η (-1)) := by sorry

end RobustOdds.SVM
Source
Tsipras et al., Robustness May Be at Odds with Accuracy, arXiv:1805.12152v5, p. 18, App. D, proof of Lemma D.5, two objective displays
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