Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§2.2, p. 796, directed analogue (§2.4) — LP duality: an optimal distance function has total weight W = f

Proved
LeightonRao.Directed.lp_duality

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

directed-graphlinear-programminglp-dualitymulticommodity-flowp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Let GGG be a strongly connected directed network on n≥2n\ge2n≥2 nodes, and let fff be the max-flow of its directed uniform multicommodity flow problem. Then:

  1. every nonnegative distance function ddd satisfying the directed distance constraint ∑(u,v)∈V2d(u,v)≥1\sum_{(u,v)\in V^2}d(u,v)\ge1∑(u,v)∈V2​d(u,v)≥1 has total weight W=∑u,vC(u,v)d(u,v)≥fW=\sum_{u,v}C(u,v)d(u,v)\ge fW=∑u,v​C(u,v)d(u,v)≥f; and
  2. some nonnegative distance function ddd satisfying the distance constraint has
W=f.W=f.W=f.

This is the linear programming duality between the concurrent flow LP and its dual (the distance LP). It reduces the lower bound of Theorem 12 to Lemma 16: a cut of small ratio cost is found from an optimal dual solution.

Formalization Note The paper states this duality in §2.2 for undirected UMFPs and uses its directed analogue in §2.4 ("The proof of Theorem 12 follows immediately from Lemma 16") without restating it. Strong connectivity and n≥2n\ge2n≥2 make the shortest-path distances well defined and the max-flow finite and positive.

Preamble
import Mathlib
import Definitions.Def_LeightonRao_Directed_Setting
Formal statement
namespace LeightonRao.Directed

/-- Leighton–Rao, §2.2 p. 796, directed analogue used in §2.4: every nonnegative distance function
satisfying the directed distance constraint has total weight at least the max-flow, and some such
distance function has total weight exactly the max-flow. -/
theorem lp_duality {V : Type} [Fintype V] [DecidableEq V] (N : DiNetwork V)
    (hn : 2 ≤ Fintype.card V) (hconn : IsStronglyConnectedNet N) :
    (∀ d : V → V → ℝ, (∀ u v, 0 ≤ d u v) → SatisfiesDiConstraint N d →
        diMaxFlow N diDemand ≤ diTotalWeight N d) ∧
    (∃ d : V → V → ℝ, (∀ u v, 0 ≤ d u v) ∧ SatisfiesDiConstraint N d ∧
        diTotalWeight N d = diMaxFlow N diDemand) := by sorry

end LeightonRao.Directed
Source
Leighton and Rao, Multicommodity max-flow min-cut theorems and their use in designing approximation algorithms, J. ACM 46 (1999), p. 796, §2.2, LP duality sentence; directed analogue used in §2.4, p. 807
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