Approximate Nash Equilibrium via Inexact ADMM

• The paper presents a distributed inexact-ADMM method to compute ε-approximate Nash equilibria in convex and strongly monotone games.
• It leverages consensus constraints and proximal updates, ensuring convergence with an O(1/k) residual decay under limited information exchange.
• The algorithm’s tuning of penalty parameters balances convergence speed and stability, with empirical validation in networked scenarios like wireless congestion control.

An approximate Nash equilibrium seeking algorithm is a computational procedure designed to identify an action profile or strategy set for multiple agents in a noncooperative game, such that no agent can achieve more than a specified ε improvement in their cost or utility by deviating unilaterally. Rigorous development of such algorithms is central to multi-agent learning, distributed optimization, and equilibrium computation for games characterized by convexity, continuity, monotonicity, and possibly large-scale communication graphs. Among the foundational approaches, distributed inexact-ADMM (Alternating Direction Method of Multipliers) provides a principled framework for convergence-guaranteed iterative computation of approximate Nash equilibria under restricted information and network constraints (Salehisadaghiani et al., 2016).

1. Precise Problem Formulation

Consider a game $\mathcal{N}=\{1,\ldots,N\}$ of $N$ players, each selecting $x_i\in X_i\subset\mathbb{R}$ (convex, compact), with joint action vector $x=(x_i,x_{-i})\in X:=\prod_j X_j$, and cost function $J_i(x_i,x_{-i})$ for player $i$. The Nash equilibrium $x^*$ is characterized by

$$J_i(x_i^*,x_{-i}^*) \leq J_i(x_i,x_{-i}^*) \quad \forall x_i\in X_i,\ \forall i,$$

which is equivalently reformulated as a variational inequality (VI) involving the pseudo-gradient mapping

$F(x):=(\nabla_1 J_1(x),\ldots,\nabla_N J_N(x))^T.$

A solution $x^*$ to

$N$0

yields the Nash equilibrium. To enable distributed computation, each agent maintains a local copy $N$1 and consensus constraints are imposed via a communication graph $N$2.

An $N$3-approximate Nash equilibrium is $N$4 such that: $N$5

2. Algorithmic Framework: Inexact-ADMM Approach

The distributed NE seeking problem under consensus constraints is framed as

$N$6

with $N$7 the indicator function for $N$8.

The edge-based augmented Lagrangian is: $N$9 where $x_i\in X_i\subset\mathbb{R}$0 are dual variables and $x_i\in X_i\subset\mathbb{R}$1 is the penalty parameter.

ADMM update steps per player $x_i\in X_i\subset\mathbb{R}$2, per iteration $x_i\in X_i\subset\mathbb{R}$3:

1. Primal update (x-step):

$x_i\in X_i\subset\mathbb{R}$4

2. Consensus update (z-step):

$x_i\in X_i\subset\mathbb{R}$5

3. Dual update (λ-step):

$x_i\in X_i\subset\mathbb{R}$6

Communication per step requires player $x_i\in X_i\subset\mathbb{R}$7 to receive $x_i\in X_i\subset\mathbb{R}$8 from each neighbor $x_i\in X_i\subset\mathbb{R}$9.

3. Convergence Guarantees and Analysis

Assuming:

• Nonempty, compact, convex $x=(x_i,x_{-i})\in X:=\prod_j X_j$0 for all $x=(x_i,x_{-i})\in X:=\prod_j X_j$1.
• $x=(x_i,x_{-i})\in X:=\prod_j X_j$2 is $x=(x_i,x_{-i})\in X:=\prod_j X_j$3 in $x=(x_i,x_{-i})\in X:=\prod_j X_j$4, convex in $x=(x_i,x_{-i})\in X:=\prod_j X_j$5, joint continuity.
• $x=(x_i,x_{-i})\in X:=\prod_j X_j$6 is $x=(x_i,x_{-i})\in X:=\prod_j X_j$7-Lipschitz and $x=(x_i,x_{-i})\in X:=\prod_j X_j$8-strongly monotone.
• $x=(x_i,x_{-i})\in X:=\prod_j X_j$9 is connected.

Main convergence properties:

• For penalty $J_i(x_i,x_{-i})$0 (where $J_i(x_i,x_{-i})$1 is the smallest eigenvalue of the Laplacian of $J_i(x_i,x_{-i})$2), the iterates converge to $J_i(x_i,x_{-i})$3, with residuals

$J_i(x_i,x_{-i})$4

satisfying $J_i(x_i,x_{-i})$5.

Proof is via:

• Proximal-ADMM firm nonexpansiveness.
• Lyapunov function combining primal and dual errors:

$J_i(x_i,x_{-i})$6

with $J_i(x_i,x_{-i})$7 for $J_i(x_i,x_{-i})$8.

• Telescoping argument leads to $J_i(x_i,x_{-i})$9, $i$0.

4. Approximation Error and Parameter Tuning

$i$1-approximate Nash equilibrium: $i$2 such that

$i$3

After $i$4 steps,

$i$5

implies, by the Lipschitz continuity of $i$6, suboptimality $i$7. To achieve $i$8-accuracy, select $i$9.

Penalty selection:

• $x^*$0 must fulfill: $x^*$1 (ensuring strong convexity of penalized subproblems).
• Practically, choose $x^*$2 to trade-off convergence speed and numerical stability.

5. Practical Implementation and Complexity

Each iteration comprises local minimization and averaging over neighbors:

• Communication per step: each node sends one vector $x^*$3, one dual vector $x^*$4; total messages per iteration $x^*$5.

Empirical results (e.g., ad-hoc wireless network congestion control with $x^*$6 nodes):

• Convergence of $x^*$7 to $x^*$8 observed within 200 iterations.
• The ADMM-based method reaches $x^*$9 accuracy in approximately 50 iterations, compared to 400 for a best-response gradient scheme.
• Residual decays as $$J_i(x_i^*,x_{-i}^*) \leq J_i(x_i,x_{-i}^*) \quad \forall x_i\in X_i,\ \forall i,$$0, consistent with theoretical prediction.

6. Structural, Spectral, and Communication Considerations

Convergence speed and approximation quality depend on:

• Graph connectivity and Laplacian spectrum.
• Degree of coupling in cost functions (condition number affects $$J_i(x_i^*,x_{-i}^*) \leq J_i(x_i,x_{-i}^*) \quad \forall x_i\in X_i,\ \forall i,$$1).
• Local computation resources (solving convex minimizations per update).

A strong monotonicity of the pseudo-gradient and convexity of cost ensures global convergence under the specified penalties, and communication graph properties critically influence step-size and rate bounds.

7. Significance and Extensions

The distributed inexact-ADMM algorithm exemplifies a scalable, provably convergent method for approximate Nash equilibrium seeking in multi-agent convex games with limited information exchange. The O(1/k) residual decay and tunable accuracy via penalty parameters provide guarantees suitable for large-scale networks and real-time scenarios. This methodology has influenced subsequent work on consensus-based splitting, operator-theoretic distributed algorithms, and robust game-theoretic computation (Salehisadaghiani et al., 2016).

In summary, approximate Nash equilibrium seeking via inexact-ADMM leverages local linearizations, consensus averaging, and primal-dual residual control to yield distributed convergence at quantifiable rates, under standard convexity, monotonicity, and graph-connectivity assumptions.