Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Theorem 3.3, p. 591 — E_{P_m(x)}[š•Œ(š’¬(h̃, x))] is at least the objective of (16)

Proved
MinimaxSLP.RhsExtremal.mixture_lower_bound

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

gaussianmoment-problemp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1probability

Under Assumptions 2, 3 (at the constant qqq) and 5, and αk≄0\alpha_k\ge 0αk​≄0, let (Vki,vki,vk0i)k,i(V_k^i,v_k^i,v_{k0}^i)_{k,i}(Vki​,vki​,vk0i​)k,i​ be a feasible point of (16) in which every block has vk0i>0v_{k0}^i>0vk0i​>0 or is zero, and let Pm(x)P_m(x)Pm​(x) be the associated mixture of normal distributions. Then h↦U(Q(h,x))h\mapsto\mathbb U(\mathcal Q(h,x))h↦U(Q(h,x)) is Pm(x)P_m(x)Pm​(x)-integrable and

EPm(x)[U(Q(h~,x))]Ā ā‰„Ā āˆ‘k=1Kāˆ‘i=1N(αkpi′vki+vk0i(βkāˆ’Ī±kpi′Tx)).\mathbb E_{P_m(x)}\big[\mathbb U(\mathcal Q(\tilde h,x))\big]\ \ge\ \sum_{k=1}^K\sum_{i=1}^N\big(\alpha_k p_i'v_k^i+v_{k0}^i(\beta_k-\alpha_k p_i'Tx)\big).EPm​(x)​[U(Q(h~,x))] ≄ k=1āˆ‘K​i=1āˆ‘N​(αk​pi′​vki​+vk0i​(βkā€‹āˆ’Ī±k​pi′​Tx)).

Applied to an optimal solution of (16), the right-hand side is Z^DD(x)\hat Z_{DD}(x)Z^DD​(x); this is the lower half of the sandwich that proves Theorem 3.3.

Formalization Note The statement is made for every feasible point with zero-or-positive blocks, not only for an optimal one; the paper's last display, with "=Z^DD(x)=\hat Z_{DD}(x)=Z^DD​(x)", is the specialization to an optimal point.

Preamble
import Mathlib
import Definitions.Def_MomentDRO_Conf_Setting
import Definitions.Def_MinimaxSLP_RhsExtremal_Model
Formal statement
namespace MinimaxSLP.RhsExtremal

open MeasureTheory Matrix

/-- Proof of Theorem 3.3, p. 591: for every feasible `s` of (16) whose blocks have
`v_k0^i > 0` or are zero, `š•Œ(š’¬(Ā·, x))` is integrable under the mixture `P_m(x)` and
`E_{P_m(x)}[š•Œ(š’¬(h̃, x))] ≄ Ī£_{k,i} (α_k p_i′v_k^i + v_k0^i (β_k āˆ’ α_k p_i′T x))`, the objective
of (16) at `s`. -/
theorem mixture_lower_bound {r d n K N : ā„•} [NeZero K]
    (W : Matrix (Fin r) (Fin d) ā„) (T : Matrix (Fin r) (Fin n) ā„) (q : Fin d → ā„)
    (α β : Fin K → ā„) (ps : Fin N → Fin r → ā„) (μ : Fin r → ā„) (Q : Matrix (Fin r) (Fin r) ā„)
    (hα : āˆ€ k, 0 ≤ α k)
    (hA2 : āˆ€ z : Fin r → ā„, ∃ w : Fin d → ā„, 0 ≤ w ∧ W *ᵄ w = z)
    (hA3 : ∃ Ļ€ : Fin r → ā„, Wįµ€ *ᵄ Ļ€ ≤ q)
    (hA5 : Set.range ps = Set.extremePoints ā„ {Ļ€ : Fin r → ā„ | Wįµ€ *ᵄ Ļ€ ≤ q})
    (x : Fin n → ā„)
    (s : Fin K → Fin N → (Matrix (Fin r) (Fin r) ā„ Ɨ (Fin r → ā„) Ɨ ā„))
    (hs : s ∈ feasible16 K N μ Q)
    (hpos : āˆ€ k i, 0 < (s k i).2.2 ∨ s k i = 0) :
    Integrable (fun h => MinimaxSLP.ObjSDP.disutility α β (MinimaxSLP.RhsSDP.Qval W T q h x)) (mixture s) ∧
    obj16 T α β ps x s ≤ ∫ h, MinimaxSLP.ObjSDP.disutility α β (MinimaxSLP.RhsSDP.Qval W T q h x) āˆ‚(mixture s) := by sorry

end MinimaxSLP.RhsExtremal
Source
Bertsimas, Doan, Natarajan & Teo, Models for Minimax Stochastic Linear Optimization Problems with Risk Aversion, Math. Oper. Res. 35(3), 2010, p. 591, proof of Theorem 3.3
Human review
  • Endorsed by Shuze Chen Ā· Oct 7, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 Ā· 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
Ā© 2026 Prove2Me