Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proposition A.3 — the adjoint of δⁱ is δⁱ*X(s) = −∑ₜ μ(t)Wⁱ(s,t)X(s,t) = −∑_{t:(s,t)∈Eⁱ} μⁱ(tⁱ)√μ⁻ⁱ(s⁻ⁱ) X(s,t)

Proved
DecompGames.Duplication.proposition_A_3

by mikedeng1 · Oct 24, 2026 · Mathlib 0df444a (Lean v4.33.1)

flowsgame-decompositionp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-strengthenedp2o-v1

Let NNN be a finite set of players with finite strategy sets SiS^iSi, S=∏iSiS=\prod_iS^iS=∏i​Si, and let μi\mu^iμi be strictly positive measures on SiS^iSi with product measure μ(s)=∏iμi(si)\mu(s)=\prod_i\mu^i(s^i)μ(s)=∏i​μi(si) and μ−i(s−i)=∏j≠iμj(sj)\mu^{-i}(s^{-i})=\prod_{j\ne i}\mu^j(s^j)μ−i(s−i)=∏j=i​μj(sj). On C0={h:S→R}C_0=\{h:S\to\mathbb R\}C0​={h:S→R} take ⟨h,f⟩0=∑sμ(s)h(s)f(s)\langle h,f\rangle_0=\sum_s\mu(s)h(s)f(s)⟨h,f⟩0​=∑s​μ(s)h(s)f(s), and on the flows C1={X:S×S→R, X(s,t)=−X(t,s)}C_1=\{X:S\times S\to\mathbb R,\ X(s,t)=-X(t,s)\}C1​={X:S×S→R, X(s,t)=−X(t,s)} take ⟨X,Y⟩1=12∑s,tμ(s)μ(t)X(s,t)Y(s,t)\langle X,Y\rangle_1=\tfrac12\sum_{s,t}\mu(s)\mu(t)X(s,t)Y(s,t)⟨X,Y⟩1​=21​∑s,t​μ(s)μ(t)X(s,t)Y(s,t). Let EiE^iEi be the iii-comparable pairs (iii is the only player with si≠tis^i\ne t^isi=ti), Wi(s,t)=1/μ−i(s−i)W^i(s,t)=1/\sqrt{\mu^{-i}(s^{-i})}Wi(s,t)=1/μ−i(s−i)​ on EiE^iEi and 000 elsewhere, and δi(h)(s,t)=Wi(s,t)(h(t)−h(s))\delta^i(h)(s,t)=W^i(s,t)(h(t)-h(s))δi(h)(s,t)=Wi(s,t)(h(t)−h(s)).

Fix a player iii and let

δi∗X(s)=−∑t∈Sμ(t) Wi(s,t) X(s,t).(A.15)\delta^{i*}X(s)=-\sum_{t\in S}\mu(t)\,W^i(s,t)\,X(s,t). \qquad (A.15)δi∗X(s)=−t∈S∑​μ(t)Wi(s,t)X(s,t).(A.15)

Then δi∗\delta^{i*}δi∗ is the adjoint of δi\delta^iδi: for every flow X∈C1X\in C_1X∈C1​, δi∗X\delta^{i*}Xδi∗X is the unique f∈C0f\in C_0f∈C0​ with ⟨δih,X⟩1=⟨h,f⟩0\langle\delta^ih,X\rangle_1=\langle h,f\rangle_0⟨δih,X⟩1​=⟨h,f⟩0​ for all h∈C0h\in C_0h∈C0​ (A.14). Moreover, for every flow XXX and every s∈Ss\in Ss∈S,

δi∗X(s)=−∑t: (s,t)∈Eiμi(ti)μ−i(s−i) X(s,t).(A.16)\delta^{i*}X(s)=-\sum_{t:\,(s,t)\in E^i}\mu^i(t^i)\sqrt{\mu^{-i}(s^{-i})}\,X(s,t). \qquad (A.16)δi∗X(s)=−t:(s,t)∈Ei∑​μi(ti)μ−i(s−i)​X(s,t).(A.16)

This is the explicit form of the adjoint gradient used to express harmonicity as D(g)∈Ker⁡δ∗D(g)\in\operatorname{Ker}\delta^*D(g)∈Kerδ∗ (Proposition A.5).

Differences from the source:

  • generalization: standing assumptions n ≥ 2 players and |Sⁱ| ≥ 2 strategies for every player (p. 7) → any finite type of players ι and any finite strategy sets S i, of any size; reason: the result holds without these assumptions (with an empty Sⁱ the space of profiles is empty and every game is 0); every instance of the source is an instance of the Lean statement
  • encoding: μⁱ a strictly positive finite measure on Sⁱ → μ i : S i → ℝ, the point masses, with 0 < μ i a for all a (IsMeasureVec); reason: a measure on a finite set is determined by its point masses; μⁱ(Sⁱ) is mass μ i = ∑ₐ μ i a
  • encoding: statements over s⁻ⁱ ∈ S⁻ⁱ and profiles (tⁱ, s⁻ⁱ) → statements over full profiles s : ∀ j, S j, with (tⁱ, s⁻ⁱ) written Function.update s i tⁱ; reason: the coordinate sⁱ of s is overwritten, so the quantifier ranges over the same subprofiles
  • encoding: X ∈ C₁, the skew-symmetric functions on S × S (A.9) → X : S → S → ℝ with the hypothesis IsFlow X (X s t = −X t s); reason: same set of flows
  • encoding: δⁱ* is the adjoint of δⁱ, the unique linear operator with ⟨δⁱh, X⟩₁ = ⟨h, δⁱ*X⟩₀ (A.14), and it satisfies (A.15) → deltaStar μ i is defined by (A.15), and for every flow X and every f : C₀, (∀ h, ⟨δⁱh, X⟩₁ = ⟨h, f⟩₀) ↔ f = deltaStar μ i X; reason: this says exactly that the adjoint, which is unique, is given by (A.15)
Preamble
import Mathlib
import Definitions.Def_DecompGames_Duplication_Operators
Formal statement
namespace DecompGames.Duplication

open Finset

open Classical in
/-- Proposition A.3, p. 27: the adjoint `δⁱ*` of `δⁱ` (for `⟨·,·⟩₁` on flows and `⟨·,·⟩₀` on `C₀`,
(A.14)) is given by (A.15), `δⁱ*X(s) = −∑_{t ∈ S} μ(t) Wⁱ(s, t) X(s, t)`, and (A.15) equals (A.16),
`−∑_{t : (s,t) ∈ Eⁱ} μⁱ(tⁱ) √(μ⁻ⁱ(s⁻ⁱ)) X(s, t)`. -/
theorem proposition_A_3 {ι : Type} [Fintype ι] [DecidableEq ι] {S : ι → Type}
    [∀ i, Fintype (S i)] [∀ i, DecidableEq (S i)]
    (μ : ∀ i, S i → ℝ) (hμ : IsMeasureVec μ) (i : ι) :
    (∀ X : (∀ j, S j) → (∀ j, S j) → ℝ, IsFlow X →
      ∀ f : (∀ j, S j) → ℝ,
        (∀ h : (∀ j, S j) → ℝ, ip1 μ (delta μ i h) X = ip0 μ h f) ↔ f = deltaStar μ i X) ∧
    (∀ X : (∀ j, S j) → (∀ j, S j) → ℝ, IsFlow X → ∀ s : ∀ j, S j,
      deltaStar μ i X s =
        -∑ t ∈ univ.filter (fun t => IsComparable i s t),
          μ i (t i) * Real.sqrt (productMeasureExcept μ i s) * X s t) := by sorry

end DecompGames.Duplication
Source
Abdou, Pnevmatikos, Scarsini & Venel, Decomposition of Games: Some Strategic Considerations, arXiv:1901.06048v2, p. 27, Proposition A.3, (A.14)–(A.16)
Human review
  • Endorsed by Shuze Chen · Oct 24, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 24, 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