Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Remark after Theorem 1, p. 910 — if r⁰ + r′ζ ⩽ 0 on 𝒱, the worst case is 0, attained in (21) at s = 0

Proved
TractableDRO.MeanSupport.remark_nonpos

by mikedeng1 · Oct 7, 2026 · Mathlib 0df444a (Lean v4.33.1)

distributionally-robust-optimizationmoment-boundsp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

This is the second case of the Remark after Theorem 1 of Goh and Sim (2010).

Let V,V^⊆Rn\mathcal V, \widehat{\mathcal V} \subseteq \mathbb R^nV,V⊆Rn with V^\widehat{\mathcal V}V nonempty and V^⊆V\widehat{\mathcal V} \subseteq \mathcal VV⊆V, and let F1\mathbb F_1F1​ be the family of distributions with integrable coordinates, mean in V^\widehat{\mathcal V}V and support in V\mathcal VV. Let r0∈Rr^0 \in \mathbb Rr0∈R, r∈Rnr \in \mathbb R^nr∈Rn satisfy r0+r′ζ≤0r^0 + r'\zeta \le 0r0+r′ζ≤0 for all ζ∈V\zeta \in \mathcal Vζ∈V. Then

sup⁡P∈F1EP((r0+r′ζ~)+)=0,\sup_{\mathbb P \in \mathbb F_1} \mathbb E_{\mathbb P}\big((r^0 + r'\tilde\zeta)^+\big) = 0,P∈F1​sup​EP​((r0+r′ζ~​)+)=0,

and the infimum defining π1(r0,r)\pi^1(r^0, r)π1(r0,r) in (21) is attained at s=0s = 0s=0: writing Φ(s)\Phi(s)Φ(s) for the objective of (21),

Φ(0)=π1(r0,r)=0.\Phi(0) = \pi^1(r^0, r) = 0.Φ(0)=π1(r0,r)=0.

When the affine function is nonpositive on the support, the positive part vanishes almost surely and both the worst case and the bound are zero.

Formalization Note The page states the implication with no condition on the sets; when F1=∅\mathbb F_1 = \emptysetF1​=∅ the left side would be −∞-\infty−∞. As in the first case of the Remark, V^⊆V\widehat{\mathcal V} \subseteq \mathcal VV⊆V (the paper's construction, §4.5) and V^≠∅\widehat{\mathcal V} \ne \emptysetV=∅ are hypotheses; together they make F1\mathbb F_1F1​ nonempty. Values are in EReal; Φ\PhiΦ is pi1Obj.

Preamble
import Mathlib
import Definitions.Def_TractableDRO_MeanSupport_Model

open MeasureTheory
Formal statement
namespace TractableDRO.MeanSupport

theorem remark_nonpos {n : ℕ} (V Vhat : Set (Fin n → ℝ)) (hsub : Vhat ⊆ V)
    (hne : Vhat.Nonempty) (r0 : ℝ) (r : Fin n → ℝ) (hnp : ∀ ζ ∈ V, r0 + r ⬝ᵥ ζ ≤ 0) :
    worstCase (family1 V Vhat) r0 r = 0 ∧
      pi1Obj V Vhat r0 r 0 = pi1 V Vhat r0 r ∧ pi1 V Vhat r0 r = 0 := by sorry

end TractableDRO.MeanSupport
Source
Goh & Sim, Distributionally Robust Optimization and Its Tractable Approximations, Oper. Res. 58(4), 2010, p. 910, Remark after Theorem 1 (second implication)
Human review
  • Endorsed by Shuze Chen · Oct 7, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 7, 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