Decoupled SGDA for Games with Intermittent Strategy Communication

2025-01-24

AI Summary

To address the challenges of high communication overhead and difficulties in synchronization—arising from noisy or delayed opponent strategies in multi-player games—this paper proposes Decoupled Stochastic Gradient Descent-Ascent (Decoupled SGDA). The method fundamentally decouples SGDA into two phases: asynchronous local updates and periodic global synchronization, enabling players to optimize independently using stale strategies and substantially reducing communication dependency. Theoretically, Decoupled SGDA achieves optimal communication complexity in strongly monotone–convex-concave (SCSC) games and significantly reduces communication costs in weakly coupled games. Its convergence is rigorously established under standard assumptions, and it demonstrates superior empirical performance over existing federated minimax methods in non-equilibrium, noisy settings.

Technology Category

• Multiagent Systems: Mechanism Design
• Game Theory and Economic Paradigms: Mechanism Design
• Machine Learning: Distributed Machine Learning & Federated Learning

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

We focus on reducing communication overhead in multiplayer games, where frequently exchanging strategies between players is not feasible and players have noisy or outdated strategies of the other players. We introduce Decoupled SGDA, a novel adaptation of Stochastic Gradient Descent Ascent (SGDA). In this approach, players independently update their strategies based on outdated opponent strategies, with periodic synchronization to align strategies. For Strongly-Convex-Strongly-Concave (SCSC) games, we demonstrate that Decoupled SGDA achieves near-optimal communication complexity comparable to the best-known GDA rates. For weakly coupled games where the interaction between players is lower relative to the non-interactive part of the game, Decoupled SGDA significantly reduces communication costs compared to standard SGDA. Our findings extend to multi-player games. To provide insights into the effect of communication frequency and convergence, we extensively study the convergence of Decoupled SGDA for quadratic minimax problems. Lastly, in settings where the noise over the players is imbalanced, Decoupled SGDA significantly outperforms federated minimax methods.

Problem

• Multi-player Games
• Strategy Synchronization
• Communication Efficiency

Innovation

• Decoupled SGDA
• Reduced Communication Cost
• Multi-player Games