Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

p. 367, proof of THEOREM 2 — G has maximum degree 4 (corrected: the degree-4 nodes are d_j / d'_j of one-sign 3-literal clauses)

Proved
EdgeDomSets.NPC.G_degree

by mikedeng1 · Oct 24, 2026 · Mathlib 0df444a (Lean v4.33.1)

degreeedge-dominating-setp2o-batch-pfp2ap2o-correctedp2o-gran-per-chapterp2o-plan-paperp2o-strengthenedp2o-v1

Let φ\varphiφ be a SAT-3-restricted formula and GGG its graph. Every vertex of GGG has degree at most 4, and a vertex www has degree 4 if and only if

w=dj for a clause Cj of three positive literals, or w=dj′ for a clause Cj of three negated literals.w=d_j\ \text{for a clause } C_j \text{ of three positive literals, or } w=d'_j \text{ for a clause } C_j \text{ of three negated literals.}w=dj​ for a clause Cj​ of three positive literals, or w=dj′​ for a clause Cj​ of three negated literals.

These are the vertices that the substitution of the gadget HHH removes, so that G∗G^*G∗ has maximum degree 3.

Formalization Note Clause and position indices are 0-based (Fin). The vertices of GGG are indexed by literal occurrences: the occurrence ooo has a literal vertex lit o (the paper's (ai,j1)(a_i,j_1)(ai​,j1​), (bi,j2)(b_i,j_2)(bi​,j2​) or (ci,j3)(c_i,j_3)(ci​,j3​)), a pendant pend o (the primed copy) and a variable-side vertex varNode o (aia_iai​ or bib_ibi​ for the two positive occurrences of xix_ixi​, cic_ici​ for the negated one); dNode j, dNode' j are djd_jdj​, dj′d'_jdj′​. "Degree at least 4" is written as the negation of DegreeLE (G φ) w 3.

Differences from the source:

  • encoding: vertices a_i, b_i, c_i, (a_i, j1), (b_i, j2), (c_i, j3), their primed copies, d_j, d'_j, with 1-based indices → one literal vertex, one pendant vertex and one variable-side vertex per literal occurrence (clause index, position), plus d_j, d'_j, all 0-based (Fin); reason: under SAT-3-restricted the three occurrences of x_i are exactly x_i ∈ C_j1, x_i ∈ C_j2, x̄_i ∈ C_j3, so the vertices and edges are the printed ones
  • correction: the degree-4 nodes are the d_i's where C_i is a 3-literal clause → the degree-4 nodes are d_j for C_j with three positive literals and d'_j for C_j with three negated literals; reason: d_j has degree 1 + (positive literals of C_j) and d'_j 1 + (negated ones); counterexample: C1 = (x1 ∨ x2 ∨ x̄1), C2 = (x1 ∨ x2), C3 = (x̄2) is SAT-3-restricted, C1 has 3 literals but deg d_1 = 3, deg d'_1 = 2; for C = (x̄1 ∨ x̄2 ∨ x̄3) the node d' has degree 4
  • reading: maximum degree 4 → every vertex has degree ≤ 4, with equality exactly at the listed nodes; reason: "maximum degree 4" is attained only when a one-sign 3-literal clause exists; the iff gives the attained case
  • generalization: a set of clauses with at most three literals per clause, every variable occurring two times and its negation once → IsSat3Restricted φ on a list of literal lists: occurrences counted with multiplicity, so a clause may repeat a literal or hold x and x̄, clauses may repeat, and empty clauses are allowed; reason: every instance of the printed problem satisfies IsSat3Restricted, so the statement specializes to the source's claim on each of them; the construction and the argument are unchanged on the extra formulas (paper.md reading 4)
Preamble
import Mathlib
import Definitions.Def_EdgeDomSets_NPC_Construction
Formal statement
namespace EdgeDomSets.NPC

theorem G_degree (φ : CookLevin.Formula) (hφ : IsSat3Restricted φ) :
    MaxDegreeLE (G φ) 4 ∧
      ∀ w : GV φ, ¬ DegreeLE (G φ) w 3 ↔
        ∃ j : Fin φ.length, (φ.get j).length = 3 ∧
          ((w = dNode j ∧ ∀ l ∈ φ.get j, ∃ v, l = CookLevin.Literal.Pos v) ∨
            (w = dNode' j ∧ ∀ l ∈ φ.get j, ∃ v, l = CookLevin.Literal.Neg v)) := by sorry

end EdgeDomSets.NPC
Source
Yannakakis, Gavril, Edge dominating sets in graphs, SIAM J. Appl. Math. 38(3) (1980) 364–372, p. 367, proof of THEOREM 2, last paragraph ("The graph G constructed above has maximum degree 4.")
Human review
  • Endorsed by Shuze Chen · Oct 24, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 24, 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