Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

(25), proof of Lemma 4.1, p. 13 — the shifted direction ηz⁽⁰⁾ + d solves the homogeneous equations, η = 1 − γ

Proved
ConicQuadIPM.NewtonStep.eq_25

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

conic-optimizationinterior-pointnewton-directionp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Let K=K1×⋯×KkK = K^1\times\cdots\times K^kK=K1×⋯×Kk be a product of cones of the kinds R+\mathbb R_+R+​, KqK^qKq, KrK^rKr with the usual dimension conventions, let A∈Rm×nA\in\mathbb R^{m\times n}A∈Rm×n, b∈Rmb\in\mathbb R^mb∈Rm, c∈Rnc\in\mathbb R^nc∈Rn, let (x(0),τ(0),y(0),s(0),κ(0))(x^{(0)},\tau^{(0)},y^{(0)},s^{(0)},\kappa^{(0)})(x(0),τ(0),y(0),s(0),κ(0)) be any point, γ∈[0,1]\gamma\in[0,1]γ∈[0,1], and let (dx,dτ,dy,ds,dκ)(d_x,d_\tau,d_y,d_s,d_\kappa)(dx​,dτ​,dy​,ds​,dκ​) solve the Newton system (22). Put η=1−γ\eta = 1-\gammaη=1−γ. Then

A(ηx(0)+dx)−b(ητ(0)+dτ)=0,AT(ηy(0)+dy)+(ηs(0)+ds)−c(ητ(0)+dτ)=0,−cT(ηx(0)+dx)+bT(ηy(0)+dy)−(ηκ(0)+dκ)=0.\begin{aligned} A(\eta x^{(0)} + d_x) - b(\eta\tau^{(0)} + d_\tau) &= 0,\\ A^T(\eta y^{(0)} + d_y) + (\eta s^{(0)} + d_s) - c(\eta\tau^{(0)} + d_\tau) &= 0,\\ -c^T(\eta x^{(0)} + d_x) + b^T(\eta y^{(0)} + d_y) - (\eta\kappa^{(0)} + d_\kappa) &= 0. \end{aligned}A(ηx(0)+dx​)−b(ητ(0)+dτ​)AT(ηy(0)+dy​)+(ηs(0)+ds​)−c(ητ(0)+dτ​)−cT(ηx(0)+dx​)+bT(ηy(0)+dy​)−(ηκ(0)+dκ​)​=0,=0,=0.​

In words: the point η(x(0),τ(0),y(0),s(0),κ(0))+(dx,dτ,dy,ds,dκ)\eta(x^{(0)},\tau^{(0)},y^{(0)},s^{(0)},\kappa^{(0)}) + (d_x,d_\tau,d_y,d_s,d_\kappa)η(x(0),τ(0),y(0),s(0),κ(0))+(dx​,dτ​,dy​,ds​,dκ​) satisfies the linear equations of the homogeneous model. This is the first step of the proof of Lemma 4.1, and it is what makes the skew-symmetry argument (26) applicable.

Formalization Note The page prints −c(ηx(0)+dx)-c(\eta x^{(0)}+d_x)−c(ηx(0)+dx​) in the third line; the formal statement uses −cT(ηx(0)+dx)-c^T(\eta x^{(0)}+d_x)−cT(ηx(0)+dx​). The second identity is stated block by block. η\etaη is written out as 1−γ1-\gamma1−γ. The hypothesis γ∈[0,1]\gamma\in[0,1]γ∈[0,1] is the range of (21) and is not needed for the identity.

Preamble
import Mathlib
import Definitions.Def_ConicQuadIPM_NewtonStep_Setting
Formal statement
namespace ConicQuadIPM.NewtonStep

open Matrix

theorem eq_25
    {m k : ℕ} (kind : Fin k → ConicQuadIPM.Complementarity.ConeKind) (n : Fin k → ℕ) (hwf : ConicQuadIPM.Complementarity.WellFormed kind n)
    (A : (i : Fin k) → Matrix (Fin m) (Fin (n i)) ℝ) (b : Fin m → ℝ)
    (c : (i : Fin k) → Fin (n i) → ℝ)
    (x0 : (i : Fin k) → Fin (n i) → ℝ) (τ0 : ℝ) (y0 : Fin m → ℝ)
    (s0 : (i : Fin k) → Fin (n i) → ℝ) (κ0 : ℝ) (γ : ℝ) (hγ : γ ∈ Set.Icc (0 : ℝ) 1)
    (dx : (i : Fin k) → Fin (n i) → ℝ) (dτ : ℝ) (dy : Fin m → ℝ)
    (ds : (i : Fin k) → Fin (n i) → ℝ) (dκ : ℝ)
    (h22 : NewtonSystem kind n A b c x0 τ0 y0 s0 κ0 γ dx dτ dy ds dκ) :
    (∑ i, A i *ᵥ ((1 - γ) • x0 i + dx i)) - ((1 - γ) * τ0 + dτ) • b = 0 ∧
    (∀ i, (A i)ᵀ *ᵥ ((1 - γ) • y0 + dy) + ((1 - γ) • s0 i + ds i)
        - ((1 - γ) * τ0 + dτ) • c i = 0) ∧
    -(∑ i, c i ⬝ᵥ ((1 - γ) • x0 i + dx i)) + b ⬝ᵥ ((1 - γ) • y0 + dy)
        - ((1 - γ) * κ0 + dκ) = 0 := by sorry

end ConicQuadIPM.NewtonStep
Source
Andersen, Roos & Terlaky, On implementing a primal-dual interior-point method for conic quadratic optimization, Math. Program. (2003); authors' preprint of 18 Dec 2000, p. 13, proof of Lemma 4.1, (25)
Human review
  • Endorsed by Shuze Chen · Oct 7, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · 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
© 2026 Prove2Me