Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Proposition 3, p. 21 — |φ_{i,j} − x_{i,j}Σ_l φ_{i,l}| < ε′(1 + Σ_l φ_{i,l}) < ε²/2cn²

Proved
FixpNash.WeakApprox.coordinatewise_bound

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

fixed-pointnash-equilibriump2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

Let Γ\GammaΓ be a finite game with constants nnn (the maximum number of pure strategies of a player), MMM (the maximum over players of the spread between their largest and smallest pure payoff) and c=1+Mc=1+Mc=1+M. Let 0<ε≤10<\varepsilon\le10<ε≤1, put

ε′=ε24c2n3,\varepsilon'=\frac{\varepsilon^2}{4c^2n^3},ε′=4c2n3ε2​,

and let xxx be a weak ε′\varepsilon'ε′-approximate fixed point of Nash's function: a mixed profile with ∣FΓ(x)−x∣∞<ε′|F_\Gamma(x)-x|_\infty<\varepsilon'∣FΓ​(x)−x∣∞​<ε′. Write φi,j(x)=max⁡{0,gi,j(x)}\varphi_{i,j}(x)=\max\{0,g_{i,j}(x)\}φi,j​(x)=max{0,gi,j​(x)}. Then for every player iii and every j∈Sij\in S_ij∈Si​,

∣φi,j(x)−xi,j∑lφi,l(x)∣<ε′(1+∑lφi,l(x))<ε22cn2,\Bigl|\varphi_{i,j}(x)-x_{i,j}\sum_{l}\varphi_{i,l}(x)\Bigr|<\varepsilon'\Bigl(1+\sum_l\varphi_{i,l}(x)\Bigr)<\frac{\varepsilon^2}{2cn^2},​φi,j​(x)−xi,j​l∑​φi,l​(x)​<ε′(1+l∑​φi,l​(x))<2cn2ε2​,

where the sums run over l∈Sil\in S_il∈Si​.

This turns the approximate fixed-point condition into a bound on each positive gain relative to the total positive gain of the player; the case analysis that concludes the proof of Proposition 3 starts from it.

Formalization Note Both inequalities of the page's chain are stated, as a conjunction. ε≤1\varepsilon\le1ε≤1 is the page's "Assume without loss of generality that ϵ≤1\epsilon\le1ϵ≤1". nnn and ccc are computed from the game (maxNumStrategies, cConst); ϵ2/2cn2\epsilon^2/2cn^2ϵ2/2cn2 is read as ε2/(2cn2)\varepsilon^2/(2cn^2)ε2/(2cn2), as the page writes it two lines later.

Preamble
import Mathlib
import Definitions.Def_agt_games
import Definitions.Def_FixpNash_WeakApprox_NashFunction
import Definitions.Def_FixpNash_WeakApprox_GameConstants
Formal statement
namespace FixpNash.WeakApprox

/-- Proof of Proposition 3, p. 21: with `ε' = ε²/(4c²n³)` and `φ_{i,j} = max{0, g_{i,j}}`, a weak
`ε'`-approximate fixed point satisfies
`|φ_{i,j}(x) - x_{i,j} Σ_l φ_{i,l}(x)| < ε'(1 + Σ_l φ_{i,l}(x)) < ε²/(2cn²)`. -/
theorem coordinatewise_bound {ι : Type*} [Fintype ι] [DecidableEq ι]
    {S : ι → Type*} [∀ i, Fintype (S i)] [∀ i, DecidableEq (S i)]
    (u : ι → (∀ i, S i) → ℝ) (ε : ℝ) (hε : 0 < ε) (hε1 : ε ≤ 1) (x : ∀ i, S i → ℝ)
    (hx : IsWeakApproxFixedPoint u x
      (ε ^ 2 / (4 * cConst u ^ 2 * (maxNumStrategies S : ℝ) ^ 3))) :
    ∀ (i : ι) (j : S i),
      |max 0 (gain u x i j) - x i j * ∑ l : S i, max 0 (gain u x i l)| <
          ε ^ 2 / (4 * cConst u ^ 2 * (maxNumStrategies S : ℝ) ^ 3) *
            (1 + ∑ l : S i, max 0 (gain u x i l)) ∧
        ε ^ 2 / (4 * cConst u ^ 2 * (maxNumStrategies S : ℝ) ^ 3) *
            (1 + ∑ l : S i, max 0 (gain u x i l)) <
          ε ^ 2 / (2 * cConst u * (maxNumStrategies S : ℝ) ^ 2) := by sorry

end FixpNash.WeakApprox
Source
Etessami & Yannakakis, On the complexity of Nash equilibria and other fixed points, author manuscript (SIAM J. Comput. 39 (2010)), proof of Proposition 3, second direction, p. 21
Human review
  • Endorsed by Shuze Chen · Oct 8, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 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