Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

(55) — max⁡λmin⁡RHβ=max⁡λmin⁡AHβ=min⁡Amax⁡λHβ\max_\lambda\min_{\mathbb R}H_\beta=\max_\lambda\min_{\mathcal A}H_\beta=\min_{\mathcal A}\max_\lambda H_\betamaxλ​minR​Hβ​=maxλ​minA​Hβ​=minA​maxλ​Hβ​

Proved
WorstCaseCVaR.Mixture.eq_55

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

cvarminimaxmixture-distributionp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

Under the standing assumptions and with A=[min⁡iα‾i∗,max⁡iα‾i∗]\mathcal A=[\min_i\underline\alpha^*_i,\max_i\overline\alpha^*_i]A=[mini​α​i∗​,maxi​αi∗​] built from the minimizer intervals of the Fβi(x,⋅)F^i_\beta(x,\cdot)Fβi​(x,⋅) as in the previous milestone, there are α0∈A\alpha_0\in\mathcal Aα0​∈A and λ0∈Λ\lambda_0\in\Lambdaλ0​∈Λ such that Hβ(x,α0,λ0)H_\beta(x,\alpha_0,\lambda_0)Hβ​(x,α0​,λ0​) is simultaneously

  1. the largest value of λ↦inf⁡α∈RHβ(x,α,λ)\lambda\mapsto\inf_{\alpha\in\mathbb R}H_\beta(x,\alpha,\lambda)λ↦infα∈R​Hβ​(x,α,λ) over Λ\LambdaΛ,
  2. the largest value of λ↦min⁡α∈AHβ(x,α,λ)\lambda\mapsto\min_{\alpha\in\mathcal A}H_\beta(x,\alpha,\lambda)λ↦minα∈A​Hβ​(x,α,λ) over Λ\LambdaΛ,
  3. the smallest value of α↦max⁡λ∈ΛHβ(x,α,λ)\alpha\mapsto\max_{\lambda\in\Lambda}H_\beta(x,\alpha,\lambda)α↦maxλ∈Λ​Hβ​(x,α,λ) over A\mathcal AA.

That is,

max⁡λ∈Λmin⁡α∈RHβ(x,α,λ)=max⁡λ∈Λmin⁡α∈AHβ(x,α,λ)=min⁡α∈Amax⁡λ∈ΛHβ(x,α,λ).\max_{\lambda\in\Lambda}\min_{\alpha\in\mathbb R}H_\beta(x,\alpha,\lambda)=\max_{\lambda\in\Lambda}\min_{\alpha\in\mathcal A}H_\beta(x,\alpha,\lambda)=\min_{\alpha\in\mathcal A}\max_{\lambda\in\Lambda}H_\beta(x,\alpha,\lambda).λ∈Λmax​α∈Rmin​Hβ​(x,α,λ)=λ∈Λmax​α∈Amin​Hβ​(x,α,λ)=α∈Amin​λ∈Λmax​Hβ​(x,α,λ).

Formalization Note The inner extrema are the real sInf/sSup of images; each is attained under the hypotheses, and the outer extrema are stated as IsGreatest/IsLeast with the common value exhibited.

Preamble
import Mathlib
import Definitions.Def_WorstCaseCVaR_Mixture_Setting

open MeasureTheory
Formal statement
namespace WorstCaseCVaR.Mixture

theorem eq_55 {l m n : ℕ} [NeZero l] (P : Fin l → Measure (Fin m → ℝ))
    [∀ i, IsProbabilityMeasure (P i)] (f : (Fin n → ℝ) → (Fin m → ℝ) → ℝ)
    (β : ℝ) (hβ0 : 0 < β) (hβ1 : β < 1) (x : Fin n → ℝ)
    (hf : ∀ i, Integrable (f x) (P i))
    (a b : Fin l → ℝ)
    (hab : ∀ i, {α : ℝ | IsMinOn (Fi P f β x i) Set.univ α} = Set.Icc (a i) (b i)) :
    ∃ α₀ ∈ Set.Icc (Finset.univ.inf' Finset.univ_nonempty a)
        (Finset.univ.sup' Finset.univ_nonempty b),
      ∃ lam₀ ∈ stdSimplex ℝ (Fin l),
        IsGreatest ((fun lam => sInf (Set.range (fun α => Hfun P f β x α lam))) ''
            stdSimplex ℝ (Fin l)) (Hfun P f β x α₀ lam₀) ∧
        IsGreatest ((fun lam => sInf ((fun α => Hfun P f β x α lam) ''
            Set.Icc (Finset.univ.inf' Finset.univ_nonempty a)
              (Finset.univ.sup' Finset.univ_nonempty b))) ''
            stdSimplex ℝ (Fin l)) (Hfun P f β x α₀ lam₀) ∧
        IsLeast ((fun α => sSup ((fun lam => Hfun P f β x α lam) '' stdSimplex ℝ (Fin l))) ''
            Set.Icc (Finset.univ.inf' Finset.univ_nonempty a)
              (Finset.univ.sup' Finset.univ_nonempty b)) (Hfun P f β x α₀ lam₀) := by sorry

end WorstCaseCVaR.Mixture
Source
Zhu & Fukushima, Worst-Case Conditional Value-at-Risk with Application to Robust Portfolio Management, Oper. Res. 57(5), 2009, p. 1166, proof of Theorem 1, Eq. (55)
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 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