Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
โ† Formalpedia

Proof of Theorem 5, p. 40 โ€” the zฬƒ-marginal of every lifted distribution lies in ๐’ซโ€ฒ

Proved
DRConvexOpt.Lifting.marginal_subset_momentSet

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

ambiguity-setdistributionally-robust-optimizationmarginal-distributionp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Let g:RPโ†’RMg : \mathbb R^P \to \mathbb R^Mg:RPโ†’RM be measurable, fโˆˆRMf \in \mathbb R^MfโˆˆRM, KโІRM\mathcal K \subseteq \mathbb R^MKโІRM a proper cone, and CiโІRP\mathcal C_i \subseteq \mathbb R^PCiโ€‹โІRP, iโˆˆIi \in \mathcal IiโˆˆI, measurable sets with bounds pโ€พi,pโ€พi\underline p_i, \overline p_ipโ€‹iโ€‹,pโ€‹iโ€‹. Then

ฮ z~PโІPโ€ฒ,\Pi_{\tilde z}\mathcal P \subseteq \mathcal P',ฮ z~โ€‹PโІPโ€ฒ,

where P\mathcal PP is the lifted ambiguity set of distributions of (z~,u~)(\tilde z, \tilde u)(z~,u~) with E[u~]=f\mathbb E[\tilde u] = fE[u~]=f, g(z~)โ‰ผKu~g(\tilde z) \preccurlyeq_{\mathcal K} \tilde ug(z~)โ‰ผKโ€‹u~ almost surely and P[z~โˆˆCi]โˆˆ[pโ€พi,pโ€พi]\mathbb P[\tilde z \in \mathcal C_i] \in [\underline p_i, \overline p_i]P[z~โˆˆCiโ€‹]โˆˆ[pโ€‹iโ€‹,pโ€‹iโ€‹], and Pโ€ฒ\mathcal P'Pโ€ฒ is the set of distributions of z~\tilde zz~ with E[g(z~)]โ‰ผKf\mathbb E[g(\tilde z)] \preccurlyeq_{\mathcal K} fE[g(z~)]โ‰ผKโ€‹f and the same probability bounds.

This is the reverse inclusion of the Lifting Theorem: the auxiliary vector u~\tilde uu~ can always be projected out without violating the conic moment bound.

Formalization Note The lifted set requires g(z~)g(\tilde z)g(z~) to be integrable (an addition to the page; see the goal theorem). Expectations and probabilities of functions of z~\tilde zz~ under P\mathbb PP and under its marginal coincide because the marginal is the pushforward by the measurable first projection.

Preamble
import Mathlib
import Definitions.Def_DRConvexOpt_Lifting_Setting
Formal statement
namespace DRConvexOpt.Lifting

open MeasureTheory

/-- Proof of Theorem 5, p. 40: ฮ _zฬƒ๐’ซ โІ ๐’ซโ€ฒ. -/
theorem marginal_subset_momentSet {nP nM nI : โ„•} (g : (Fin nP โ†’ โ„) โ†’ (Fin nM โ†’ โ„))
    (hg : Measurable g) (f : Fin nM โ†’ โ„) (K : Set (Fin nM โ†’ โ„)) (hK : DRConvexOpt.Reform.IsProperCone K)
    (C : Fin nI โ†’ Set (Fin nP โ†’ โ„)) (hC : โˆ€ i, MeasurableSet (C i)) (plo phi : Fin nI โ†’ โ„) :
    marginal (liftedSet g f K C plo phi) โІ momentSet g f K C plo phi := by sorry

end DRConvexOpt.Lifting
Source
Wiesemann, Kuhn & Sim, Distributionally Robust Convex Optimization, Optimization Online preprint 3757 (version of September 22, 2013), p. 40, proof of Theorem 5, second inclusion ๐’ซโ€ฒ โЇ ฮ _zฬƒ๐’ซ
Human review
  • Endorsed by Shuze Chen ยท Oct 9, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 ยท 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
ยฉ 2026 Prove2Me