Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proposition 3.4, p. 11, first bullet, for G the expected reward of a proper scoring rule and q ∈ D — Bregman divergence is non-negative

Proved
QAPooling.MaxMin.proposition_3_4

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

bregman-divergencep2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1proper-scoring-rule

Let sss be a regular proper scoring rule with expected reward function GGG, forecast domain D\mathcal DD and exposure function ggg, and write DG(x ∥ q)=G(x)−G(q)−⟨g(q),x−q⟩D_G(x\,\|\,q)=G(x)-G(q)-\langle g(q),x-q\rangleDG​(x∥q)=G(x)−G(q)−⟨g(q),x−q⟩ for the Bregman divergence. For every q∈Dq\in\mathcal Dq∈D and every x∈Δnx\in\Delta^nx∈Δn,

DG(x ∥ q) ≥ 0,D_G(x\,\|\,q)\ \ge\ 0,DG​(x∥q) ≥ 0,

with equality only when x=qx=qx=q.

By Remark 3.3 of the paper, DG(x ∥ q)D_G(x\,\|\,q)DG​(x∥q) is the expected reward an expert with belief xxx loses by reporting qqq. Non-negativity and strict positivity off the diagonal are what make the agent's worst-case profit in Theorem 4.1 non-negative, and positive unless all experts with positive weight agree.

Formalization Note The paper states this for a general differentiable strictly convex GGG; here GGG is the expected reward of a proper scoring rule and an exposure function plays the role of the gradient, with the second argument qqq in D\mathcal DD (where g(q)g(q)g(q) is a finite subgradient). The second bullet of Proposition 3.4 (strict convexity of DG(⋅ ∥ q)D_G(\cdot\,\|\,q)DG​(⋅∥q)) is not formalized.

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

/-- Proposition 3.4, first bullet (p. 11), for `G` the expected reward of a regular proper scoring
rule and `g` an exposure function: for `q ∈ D` and `x ∈ Δⁿ`, `D_G(x ∥ q) ≥ 0`, with equality only
when `x = q`. -/
theorem proposition_3_4 {n : ℕ} (s : ScoringRule n) (g : (Fin n → ℝ) → Fin n → ℝ)
    (hreg : IsRegular s) (hprop : IsProper s) (hg : IsExposure s g)
    (q : Fin n → ℝ) (hq : q ∈ forecastDomain s) (x : Fin n → ℝ) (hx : x ∈ stdSimplex ℝ (Fin n)) :
    0 ≤ bregman s g x q ∧ (bregman s g x q = 0 → x = q) := by sorry

end QAPooling.MaxMin
Source
Neyman & Roughgarden, From Proper Scoring Rules to Max-Min Optimal Forecast Aggregation, arXiv:2102.07081v2, p. 11, Proposition 3.4 (first bullet), Definition 3.2
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