Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 7.33 -- polyhedral_l_optimality_criterion

Proved
DiscreteConvex.LConvexFunctionsC.polyhedral_l_optimality_criterion

by Shuze Chen · Sep 28, 2026 · Mathlib 0df444a (Lean v4.33.1)

convex-optimizationdiscrete-convex-analysis

Theorem 7.33 (L-optimality criterion; p.193). GOAL. (1) For a polyhedral L-convex function g∈L[R→R]g\in L[\mathbb R\to\mathbb R]g∈L[R→R] and p∈dom⁡Rgp\in\operatorname{dom}_{\mathbb R} gp∈domR​g: g(p)≤g(q)g(p)\le g(q)g(p)≤g(q) for all qqq iff g′(p;χY)≥0g'(p;\chi_Y)\ge 0g′(p;χY​)≥0 for all Y⊆VY\subseteq VY⊆V and g′(p;1)=0g'(p;\mathbf 1)=0g′(p;1)=0. (2) For a polyhedral L♮^\natural♮-convex function g∈L♮[R→R]g\in L^\natural[\mathbb R\to\mathbb R]g∈L♮[R→R] and p∈dom⁡Rgp\in\operatorname{dom}_{\mathbb R} gp∈domR​g: g(p)≤g(q)g(p)\le g(q)g(p)≤g(q) for all qqq iff g′(p;±χY)≥0g'(p;\pm\chi_Y)\ge 0g′(p;±χY​)≥0 for all Y⊆VY\subseteq VY⊆V.

The polyhedral analogue of the L-optimality criterion (Theorem 7.14, formalized in mission 08-lconvex-functions-i) and the direct L-side mirror of mission 24-ch06d-mconvexfunctions's M-optimality criterion (Theorem 6.52): global optimality of a polyhedral L-(natural-)convex function reduces to checking finitely many directional derivatives.

(Murota, Discrete Convex Analysis, SIAM 2003, DOI 10.1137/1.9780898718508, p.193, Theorem 7.33.)

Preamble
import Mathlib
import Definitions.Def_DiscreteConvex_LConvexFunctionsC_DomR
import Definitions.Def_DiscreteConvex_LConvexFunctionsC_SBFR
import Definitions.Def_DiscreteConvex_LConvexFunctionsC_TRFR
import Definitions.Def_DiscreteConvex_LConvexFunctionsC_LNaturalConvexR
import Definitions.Def_DiscreteConvex_LConvexFunctionsC_DirDeriv
import Definitions.Def_DiscreteConvex_LConvexFunctionsC_IsPolyhedralConvex
Formal statement
namespace DiscreteConvex.LConvexFunctionsC

open Classical
open scoped Pointwise
variable {V : Type*} [Fintype V] [DecidableEq V]
/-- Theorem 7.33 (L-optimality criterion; p.193). GOAL. Global optimality of a polyhedral L-convex
or L♮-convex function is characterized by directional derivatives along coordinate-subset
indicators. Polyhedrality is the book's standing hypothesis and is not implied by (SBF[R]) and
(TRF[R]): `DirDeriv` is a `WithTop` infimum of difference quotients and reads the junk value `0`
when they are unbounded below, which `g(p) = -√(1 - (p₁ - p₂)²)` exploits at `p = (-1, 0)`. -/
theorem polyhedral_l_optimality_criterion (g : (V → ℝ) → WithTop ℝ)
    (hpoly : IsPolyhedralConvex g) (p : V → ℝ) (hp : p ∈ DomR g) :
    ((SBFR g ∧ TRFR g) →
      ((∀ q, g p ≤ g q) ↔
        ((∀ Y : Finset V, DirDeriv g p (fun v => if v ∈ Y then (1 : ℝ) else 0) ≥ 0) ∧
          DirDeriv g p (fun _ => (1 : ℝ)) = 0))) ∧
    (LNaturalConvexR g →
      ((∀ q, g p ≤ g q) ↔
        (∀ Y : Finset V, DirDeriv g p (fun v => if v ∈ Y then (1 : ℝ) else 0) ≥ 0 ∧
          DirDeriv g p (fun v => if v ∈ Y then (-1 : ℝ) else 0) ≥ 0))) := by sorry

end DiscreteConvex.LConvexFunctionsC
Source
Murota, Discrete Convex Analysis, SIAM 2003, DOI 10.1137/1.9780898718508, p.193, Theorem 7.33
Human review
  • Endorsed by Community (Bot) · Oct 1, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Shuze Chen · Oct 1, 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