Equilibria in multiagent online problems with predictions

2024-05-20

arXiv.org

AI Summary

This paper investigates how predictive capability affects equilibrium properties and social efficiency in multi-agent online games, using a collaborative ski rental problem—where agents may either jointly purchase a perpetual license or rent daily—as a canonical model. Method: We introduce, for the first time, an explicit prediction mechanism into the multi-agent online game framework, yielding a prediction-augmented game model that integrates online algorithm analysis, game-theoretic modeling, and competitive ratio theory. Contribution/Results: We rigorously characterize the existence and stability boundaries of Nash equilibria under prediction error; prove that predictions can significantly improve the collective competitive ratio; and design a distributed protocol achieving near-social-optimality under bounded prediction accuracy. Our core contribution lies in revealing how “prediction–response” interactions engender novel equilibrium structures, thereby establishing a theoretical foundation for prediction-enabled coordinated decision-making.

Technology Category

• Multiagent Systems: Mechanism Design
• Game Theory and Economic Paradigms: Coordination and Collaboration
• Machine Learning: Online Learning & Bandits

Application Category

• Economics, Online Markets and Human Computation: The sharing economy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

We study the power of (competitive) algorithms with predictions in a multiagent setting. To this goal, we introduce a multiagent version of the ski-rental problem. In this problem agents can collaborate by pooling resources to get a group license for some asset. If the license price is not met then agents have to rent the asset individually for the day at a unit price. Otherwise the license becomes available forever to everyone at no extra cost. We investigate the effect of using predictors for self and others' behavior in such a setting, as well as the new equilibria formed in this way.

Problem

• Predictive Function
• Multiplayer Online Games
• Strategy Influence

Innovation

• Predictive Strategies
• Multiplayer Online Games
• Equilibrium Analysis