Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Sect. 4d — every correlated equilibrium arises from a Bayes-rational information system

Proved
Aumann1987.BayesRational.converse

by mikedeng1 · Oct 4, 2026 · Mathlib 0df444a (Lean v4.33.1)

bayesian-rationalitycorrelated-equilibriump2o-batch-pfp1ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Let f:Γ→Sf:\Gamma\to Sf:Γ→S be a correlated equilibrium on a finite probability space (Γ,q)(\Gamma,q)(Γ,q) of a game with payoffs hhh. Then there is an information system (Ω,p,(Pi),s)(\Omega,p,(\mathcal P^i),\mathbf s)(Ω,p,(Pi),s) — finite Ω\OmegaΩ, common prior ppp, partitions Pi\mathcal P^iPi, each si\mathbf s^isi measurable with respect to Pi\mathcal P^iPi — in which every player is Bayes rational at every state and

p{ω: s(ω)=a}=q{γ: f(γ)=a}for every a∈S.p\{\omega:\ \mathbf s(\omega)=a\}=q\{\gamma:\ f(\gamma)=a\}\qquad\text{for every }a\in S.p{ω: s(ω)=a}=q{γ: f(γ)=a}for every a∈S.

Together with the Main Theorem this shows that, under Bayesian rationality, information systems correspond precisely to correlated equilibria.

Formalization Note. Γ\GammaΓ and the witness Ω\OmegaΩ live in Type (universe 0).

Preamble
import Mathlib
import Definitions.Def_agt_games
import Definitions.Def_Aumann1987_BayesRational_CorrelatedEquilibrium
import Definitions.Def_Aumann1987_BayesRational_InformationSystem
Formal statement
namespace Aumann1987.BayesRational

/-- **Sect. 4d, the converse** (Aumann 1987, Econometrica 55, Sect. 4d, p. 11, PDF p. 12): "More
precisely, for each game `G` and each correlated equilibrium `f` of `G`, there is an information
system for which it is Bayes rational for the players to play in accordance with `s`, and the
resulting distribution is the same as that of `f`."

For every correlated equilibrium `f` on a finite probability space `(Γ, q)` there are a finite set
`Ω` and an information system `I` on `Ω` (common prior, partitions, action functions measurable
w.r.t. the partitions) in which every player is Bayes rational at every state and the distribution
of the action `n`-tuple `s` equals the distribution of `f`.

**Formalization Note.** `Γ` and the witness `Ω` live in `Type`, as in `IsCED`. -/
theorem converse {ι : Type*} [Fintype ι] [DecidableEq ι] {S : ι → Type*}
    {Γ : Type} [Fintype Γ] (h : ι → (∀ i, S i) → ℝ) (q : Γ → ℝ) (hq : AGT.IsLottery q)
    (f : Γ → ∀ i, S i) (hf : IsCorrelatedEquilibrium h q f) :
    ∃ (Ω : Type) (_ : Fintype Ω) (I : InformationSystem Ω S),
      (∀ i ω, IsBayesRationalAt h I.p I.P I.s i ω) ∧ distr I.p I.s = distr q f := by sorry

end Aumann1987.BayesRational
Source
Aumann, Correlated Equilibrium as an Expression of Bayesian Rationality, Econometrica 55 (1987), DOI 10.2307/1911154, p. 11 (PDF p. 12), Sect. 4d
Human review
  • Endorsed by Shuze Chen · Oct 4, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 4, 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