Theorem 2.6, p. 7 — Nash equilibria equal the C-minimizers
ProvedNashVecOpt.SharedConstraint.theorem_2_6For an -player shared-constraint game with , let be arbitrary real-valued cost functions, any shared feasible set of joint strategies, , and the ordering cone of equation (6). The whole Nash equilibrium set is exactly the vector minimizer set:
The equality applies without regularity, convexity, or existence assumptions and can therefore describe an empty equilibrium set or one with many points.
Formalization Note A joint strategy is a dependent function over a finite player type; its th unilateral deviation is Function.update. The shared feasible set is never replaced by a product set. The hypothesis reproduces the paper's standing model assumption, although the equality itself does not need it.
import Mathlib import Definitions.Def_NashVecOpt_SharedConstraint_Model
namespace NashVecOpt.SharedConstraint
variable {ι : Type} [Fintype ι] [DecidableEq ι]
{E : ι → Type} [∀ i, AddCommGroup (E i)] [∀ i, Module ℝ (E i)]
/-- Theorem 2.6, p. 7: the Nash equilibria of a shared-constraint game are exactly the
Pareto minimizers of its augmented objective under cone (6). -/
theorem theorem_2_6 (hN : 2 ≤ Fintype.card ι) (f : ι → (∀ j, E j) → ℝ)
(X : Set (∀ j, E j)) :
NE f X = vArgMin (orderCone (E := E)) (obj f) X := by sorry
end NashVecOpt.SharedConstraint
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.