Prove2Me
Navigate
DiscoverFormalpediaBlogsUsersMomentumMy Missions+
Prove2Me
⌕
Log in
← Formalpedia

Theorem 2.4 — a proper KKK-lift gives a KKK-factorization of SCS_CSC​, and a KKK-factorization gives a KKK-lift

Proved
ConeLifts.Factorization.factorization_theorem

by mikedeng1 · Sep 27, 2026 · Mathlib 0df444a (Lean v4.33.1)

cone-factorizationcone-liftsp2o-batch-p200ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Let K⊆RmK \subseteq \mathbb R^mK⊆Rm be a full-dimensional closed convex cone and C⊆RnC \subseteq \mathbb R^nC⊆Rn, n≥1n \ge 1n≥1, a convex body (compact, convex, 0∈int⁡C0 \in \operatorname{int} C0∈intC). Let SC(x,y)=1−⟨x,y⟩S_C(x,y) = 1 - \langle x, y\rangleSC​(x,y)=1−⟨x,y⟩ on ext⁡(C)×ext⁡(C∘)\operatorname{ext}(C) \times \operatorname{ext}(C^\circ)ext(C)×ext(C∘) be the slack operator of CCC. Then:

  1. if CCC has a proper KKK-lift, that is C=π(K∩L)C = \pi(K \cap L)C=π(K∩L) for an affine subspace LLL meeting int⁡K\operatorname{int} KintK and a linear map π\piπ, then SCS_CSC​ is KKK-factorizable;
  2. conversely, if SCS_CSC​ is KKK-factorizable, that is, for some maps A:ext⁡(C)→KA : \operatorname{ext}(C) \to KA:ext(C)→K and B:ext⁡(C∘)→K∗B : \operatorname{ext}(C^\circ) \to K^*B:ext(C∘)→K∗
1−⟨x,y⟩=⟨A(x),B(y)⟩for all (x,y)∈ext⁡(C)×ext⁡(C∘),1 - \langle x, y\rangle = \langle A(x), B(y)\rangle \quad \text{for all } (x,y)\in \operatorname{ext}(C)\times\operatorname{ext}(C^\circ),1−⟨x,y⟩=⟨A(x),B(y)⟩for all (x,y)∈ext(C)×ext(C∘),

then CCC has a KKK-lift (not necessarily proper).

This is the paper's central correspondence between the geometry of lifts and the algebra of cone factorizations. It extends Yannakakis' theorem, which relates polyhedral lifts of a polytope to nonnegative factorizations of its slack matrix, to arbitrary closed convex cones, in particular to positive semidefinite lifts.

Formalization Note The two halves are a conjunction of implications, not an equivalence: properness is assumed only in the first, and the lift produced by the second may be improper. n≥1n \ge 1n≥1 is the paper's implicit full-dimensionality of CCC; for n=0n = 0n=0 the first half is false (C={0}C = \{0\}C={0}, K=RmK = \mathbb R^mK=Rm, K∗={0}K^* = \{0\}K∗={0}). Rk\mathbb R^kRk is EuclideanSpace ℝ (Fin k), the polar is one-sided, and A,BA, BA,B are total functions constrained on the extreme points only.

Preamble
import Mathlib
import Definitions.Def_ConeLifts_Factorization_IsConvexBody
import Definitions.Def_ConeLifts_Factorization_IsClosedConvexCone
import Definitions.Def_ConeLifts_Factorization_HasLift
import Definitions.Def_ConeLifts_Factorization_HasProperLift
import Definitions.Def_ConeLifts_Factorization_SlackFactorizable
Formal statement
namespace ConeLifts.Factorization

/-- **Theorem 2.4** of Gouveia, Parrilo & Thomas, *Lifts of Convex Sets and Cone Factorizations*,
arXiv:1111.3164v2, p. 4: "If C has a proper K-lift then S_C is K-factorizable. Conversely, if S_C
is K-factorizable then C has a K-lift."

Standing hypotheses (Definition 2.1, p. 3): `K ⊆ ℝᵐ` is a full-dimensional closed convex cone and
`C ⊆ ℝⁿ` is a convex body (compact, convex, `0 ∈ int C`). `1 ≤ n` is the paper's implicit
"full-dimensional convex body in ℝⁿ": for `n = 0` the first half is false. The converse
concludes a `K`-lift that need not be proper. -/
theorem factorization_theorem {n m : ℕ} (hn : 1 ≤ n)
    (C : Set (EuclideanSpace ℝ (Fin n))) (hC : IsConvexBody C)
    (K : Set (EuclideanSpace ℝ (Fin m))) (hK : IsClosedConvexCone K)
    (hKint : (interior K).Nonempty) :
    (HasProperLift K C → SlackFactorizable K C) ∧ (SlackFactorizable K C → HasLift K C) := by sorry

end ConeLifts.Factorization
Source
Gouveia, Parrilo & Thomas, Lifts of Convex Sets and Cone Factorizations, arXiv:1111.3164v2, p. 4, Theorem 2.4
Human review
  • Endorsed by Shuze Chen · Sep 27, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Sep 27, 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 paper
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me