Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 2 — positive table and no Pareto optimal α-EFX allocation above 2/3 + 1/K

Proved
ParetoEFX296.Approx.theorem_2

by mikedeng1 · Nov 8, 2026 · Mathlib 0df444a (Lean v4.33.1)

efxfair-divisionp2o-batch-stoc27ap2o-gran-per-chapterp2o-plan-paperp2o-strengthenedp2o-v1pareto-optimality

For each integer K≥6K\ge6K≥6, consider the four-agent, nine-good additive table inst(K)\mathrm{inst}(K)inst(K) of Theorem 2. Every entry is positive, and for every real α>2/3+1/K\alpha>2/3+1/Kα>2/3+1/K,

(∀i,g, vi(g)>0)and∄A: A is α-EFX and Pareto optimal.\bigl(\forall i,g,\ v_i(g)>0\bigr)\quad\text{and}\quad \nexists A:\ A\text{ is }\alpha\text{-EFX and Pareto optimal}.(∀i,g, vi​(g)>0)and∄A: A is α-EFX and Pareto optimal.

Allocations are all complete integral partitions of the nine goods, possibly with empty bundles, and Pareto optimality compares with every such allocation. Some agent must strictly improve in a Pareto domination. The statement preserves all values in the explicit table and the full range K≥6K\ge6K≥6; it is the page 9 theorem together with positivity from its page 3 formulation. Taking KKK arbitrarily large recovers the impossibility for every α>2/3\alpha>2/3α>2/3. It asserts incompatibility with Pareto optimality, and does not assert nonexistence of EFX allocations alone.

Differences from the source:

  • encoding: partitions of nine goods among agents 1,2,3,4 with integer K ≥ 6 → owner maps Good → Agent with named constructors, K : ℕ ≥ 6, and real table sums; reason: bijection with all complete integral allocations and all source integers K ≥ 6; real casts preserve every utility
  • generalization: Definition 1 restricts α to (0,1] → the same distinct-agent EFX inequality is used for every real α above 2/3 + 1/K; reason: it agrees on the source range; for α > 1 positivity implies 1-EFX, and 1 > 2/3 + 1/K for K ≥ 6
Preamble
import Mathlib
import Definitions.Def_ParetoEFX296_Approx_Instance
Formal statement
namespace ParetoEFX296.Approx

theorem theorem_2 (K : ℕ) (hK : 6 ≤ K) :
    (∀ (i : Agent) (g : Good), 0 < inst K i g) ∧
    ∀ α : ℝ, (2 : ℝ) / 3 + 1 / (K : ℝ) < α →
      ¬ ∃ A : Allocation Agent Good,
        IsAlphaEFX (inst K) α A ∧ IsParetoOptimal (inst K) A := by sorry

end ParetoEFX296.Approx
Source
Sepideh Mahabadi, Prasanna Ramakrishnan, Jakub Tarnawski, Pareto Inefficiency of EFX, STOC 2027 submission #296 (October 2026), source.pdf SHA-256 5f34a4a4b942d95889f3403900ad796cd81a27e83d783e5ab04012fab5385ee1, p.9, Theorem 2; p.3, Theorem 2 positivity clause; https://github.com/sigact-stoc/stoc2027/releases/download/papers-1/296.pdf
Human review
  • Endorsed by Shuze Chen · Nov 8, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Nov 8, 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