[Paper Review] Multiple Nash-equilibrium in Quantum Game

G. N. Parfionov|ArXiv.org|Jun 6, 2008

Quantum Mechanics and Applications7 references3 citations

TL;DR

This paper analyzes the existence, uniqueness, and multiplicity of Nash equilibria in a two-player quantum game with entangled strategies. Using analytical methods on a reduced quantum game model, it establishes conditions for multiple equilibria through matrix eigenstructure and payoff parameter constraints, showing that non-degenerate games can exhibit up to two distinct eigenequilibria under specific symmetry and norm conditions on payoff coefficients.

ABSTRACT

Methods of exploring Nash equilibrium in quantum games are studied. Analytical conditions of the existence, the uniqueness or the multiplicity of the equilibria are found.

Motivation & Objective

• To analytically determine conditions for the existence, uniqueness, and multiplicity of Nash equilibria in a two-player quantum game.
• To investigate how quantum strategies—particularly non-commuting projectors and entanglement parameters—lead to multiple equilibria.
• To extend prior numerical studies by deriving closed-form criteria for equilibrium existence using linear algebra and spectral theory.
• To clarify the role of payoff coefficients and angular parameters (θ, τ) in determining equilibrium structure, especially in non-degenerate games.

Proposed method

• Reduces the quantum game to a classical game on a torus by mapping quantum strategies to unit vectors on a circle using rotation parameters θ and τ.
• Represents payoffs via a bilinear form g(x,y) = -⟨x, Ay⟩ + ⟨x, u⟩ - ⟨v, y⟩, where A = M†CM, M = Mθ, and C is a diagonal payoff matrix.
• Applies the equilibrium criterion: (x,y) is a Nash equilibrium iff -Ay + u = λx and A†x + v = μy for non-negative λ, μ.
• Uses eigenequilibrium analysis by studying the block matrix 𝒜 = [0 A; A† 0], showing that equilibria correspond to eigenvectors of 𝒜.
• Derives existence theorems by requiring ω to be an eigenvector of CMM† and imposing norm constraints on ⟨Az, z⟩ and |z|.
• Employs rectangular property of antagonistic games to prove uniqueness when ⟨Az, z⟩ ≠ |z|³, and multiplicity when equality holds.

Experimental results

Research questions

• RQ1 Under what conditions does a quantum game admit more than one Nash equilibrium?
• RQ2 How do the payoff coefficients {cj} and entanglement parameters θ, τ jointly determine the number of equilibria?
• RQ3 What role does the vector ω = b - a play in the existence and multiplicity of equilibria?
• RQ4 Can eigenequilibria exist in non-degenerate quantum games, and if so, under what symmetry conditions?
• RQ5 What is the relationship between the norm of the vector z = M†ω and the number of equilibria?

Key findings

• A non-degenerate quantum game has a unique Nash equilibrium if ⟨Az, z⟩ ≠ |z|³, where z = M†ω.
• When ⟨Az, z⟩ = |z|³, the game admits exactly two distinct eigenequilibria: (x,y) = (z/|z|, z/|z|) and (x,y) = (-z/|z|, z/|z|).
• Eigenequilibria exist only if ω is an eigenvector of the matrix CMM†, linking equilibrium existence to spectral properties of payoff and entanglement parameters.
• The angular parameters θ and τ must be equal for eigenequilibria to exist in non-degenerate games, with their value fully determined by the payoff coefficients via cos 2θ = (m−n)ω₁ω₂ / Δ.
• The equilibrium criterion is satisfied when the Lagrange multipliers λ and μ are non-negative, which holds when the norm condition ⟨Az, z⟩ ≤ |z|³ is met.
• The game’s value remains constant across all equilibria due to the rectangular property of antagonistic games, ensuring consistency in payoff outcomes.