Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Equations (4)–(5) — duality measure and centrality along a step

Proved
AdaptiveStepIPM.PredCorr.equations_4_5

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

interior-point-methodslinear-programmingp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

Let n≥1n\ge1n≥1, let x,sx,sx,s be strictly positive, and let (dx,dy,ds)(d_x,d_y,d_s)(dx​,dy​,ds​) solve the primal-dual direction system (2) at (x,s)(x,s)(x,s) with parameter γ\gammaγ. For any real step θ\thetaθ, write x(θ)=x+θdxx(\theta)=x+\theta d_xx(θ)=x+θdx​ and s(θ)=s+θdss(\theta)=s+\theta d_ss(θ)=s+θds​. Then

μ(θ)=(1−θ)μ+θγμ,X(θ)s(θ)−μ(θ)e=(1−θ)(Xs−μe)+θ2Dxds.\mu(\theta)=(1-\theta)\mu+\theta\gamma\mu,\qquad X(\theta)s(\theta)-\mu(\theta)e=(1-\theta)(Xs-\mu e)+\theta^2D_xd_s.μ(θ)=(1−θ)μ+θγμ,X(θ)s(θ)−μ(θ)e=(1−θ)(Xs−μe)+θ2Dx​ds​.

Here Dx=diag⁡(dx)D_x=\operatorname{diag}(d_x)Dx​=diag(dx​), while X(θ)=diag⁡(x(θ))X(\theta)=\operatorname{diag}(x(\theta))X(θ)=diag(x(θ)). The identities expose the second-order term that controls both algorithmic steps.

Preamble
import Mathlib
import Definitions.Def_AdaptiveStepIPM_PredCorr_Direction

open Matrix
Formal statement
namespace AdaptiveStepIPM.PredCorr

/-- Equations (4) and (5), printed p. 4. -/
theorem equations_4_5 {m n : ℕ} (hn : 1 ≤ n)
    (A : Matrix (Fin m) (Fin n) ℝ) (x s dx ds : Fin n → ℝ)
    (dy : Fin m → ℝ) (γ θ : ℝ)
    (hx : ∀ j, 0 < x j) (hs : ∀ j, 0 < s j)
    (hd : SearchDirection A x s γ dx dy ds) :
    mu (stepX x dx θ) (stepS s ds θ) =
      (1 - θ) * mu x s + θ * γ * mu x s ∧
    (fun j => stepX x dx θ j * stepS s ds θ j -
      mu (stepX x dx θ) (stepS s ds θ)) =
      (fun j => (1 - θ) * (x j * s j - mu x s) + θ ^ 2 * (dx j * ds j)) := by sorry

end AdaptiveStepIPM.PredCorr
Source
Mizuno, Todd, Ye, On adaptive-step primal-dual interior-point algorithms for linear programming, Cornell ORIE Technical Report No. 944 (1990, rev. 1991), p. 4, equations (4)–(5)
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