Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Theorem 4.1, pp. 14–15 — u(p*; j) = Σᵢ wᵢ D_G(p* ∥ pᵢ) + µⱼ

Proved
QAPooling.MaxMin.profit_identity

by mikedeng1 · Oct 22, 2026 · Mathlib 0df444a (Lean v4.33.1)

bregman-divergencep2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1proper-scoring-rulequasi-arithmetic-pooling

Let sss be a scoring rule with exposure function ggg, let p1,…,pm∈Dp_1,\dots,p_m\in\mathcal Dp1​,…,pm​∈D with weights wiw_iwi​ adding to 111, and let p∗∈Dp^*\in\mathcal Dp∗∈D. Suppose that

g(p∗)=∑i=1mwi g(pi)+λ1n+μg(p^*)=\sum_{i=1}^m w_i\,g(p_i)+\lambda\mathbf 1_n+\mug(p∗)=i=1∑m​wi​g(pi​)+λ1n​+μ

for some λ∈R\lambda\in\mathbb Rλ∈R and μ∈Rn\mu\in\mathbb R^nμ∈Rn with μj=0\mu_j=0μj​=0 whenever pj∗>0p^*_j>0pj∗​>0. Then for every outcome jjj, the agent's profit u(p∗;j)=s(p∗;j)−∑iwis(pi;j)u(p^*;j)=s(p^*;j)-\sum_i w_i s(p_i;j)u(p∗;j)=s(p∗;j)−∑i​wi​s(pi​;j) satisfies

u(p∗;j)=∑i=1mwi DG(p∗ ∥ pi)+μj.u(p^*;j)=\sum_{i=1}^m w_i\,D_G(p^*\,\|\,p_i)+\mu_j.u(p∗;j)=i=1∑m​wi​DG​(p∗∥pi​)+μj​.

This is the computation at the heart of the proof of Theorem 4.1: the profit at the QA pool is the same weighted Bregman sum for every outcome, plus a non-negative correction μj\mu_jμj​ that vanishes on the support of p∗p^*p∗.

Formalization Note The identity is algebraic: it needs neither properness nor μ≥0\mu\ge 0μ≥0, only Equation (2) at p∗p^*p∗ and at each pip_ipi​ and ∑iwi=1\sum_i w_i=1∑i​wi​=1.

Preamble
import Mathlib
import Definitions.Def_QAPooling_MaxMin_Model
Formal statement
namespace QAPooling.MaxMin

/-- The profit identity in the proof of Theorem 4.1 (pp. 14–15): if `p* ∈ D` and
`g(p*) = Σ_i w_i g(p_i) + λ 1_n + μ` with `μ_j = 0` whenever `p*_j > 0`, then for every outcome `j`,
`u(p*; j) = Σ_i w_i D_G(p* ∥ p_i) + μ_j`. -/
theorem profit_identity {n m : ℕ} (s : ScoringRule n) (g : (Fin n → ℝ) → Fin n → ℝ)
    (hg : IsExposure s g)
    (p : Fin m → Fin n → ℝ) (hp : ∀ i, p i ∈ forecastDomain s)
    (w : Fin m → ℝ) (hw1 : ∑ i, w i = 1)
    (pstar : Fin n → ℝ) (hD : pstar ∈ forecastDomain s)
    (lam : ℝ) (mu : Fin n → ℝ)
    (hkkt : ∀ j, g pstar j = ∑ i, w i * g (p i) j + lam + mu j)
    (hcs : ∀ j, 0 < pstar j → mu j = 0) :
    ∀ j, profit s p w pstar j = ((∑ i, w i * bregman s g pstar (p i) + mu j : ℝ) : EReal) := by sorry

end QAPooling.MaxMin
Source
Neyman & Roughgarden, From Proper Scoring Rules to Max-Min Optimal Forecast Aggregation, arXiv:2102.07081v2, pp. 14–15, proof of Theorem 4.1 (chain of equalities for u(p*; j))
Human review
  • Endorsed by Shuze Chen · Oct 22, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 22, 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