Learning Bilateral Team Formation in Cooperative Multi-Agent Reinforcement Learning

2025-06-24

AI Summary

Existing MARL research predominantly assumes fixed teams or unidirectional grouping, failing to address bidirectional collaboration requirements in open, dynamic environments. To bridge this gap, we propose the first bilateral team formation learning framework for dynamic populations, modeling team construction as a bidirectional matching game among agents. Our approach enforces symmetry and reciprocity constraints to guide collaborative evolution. Integrating multi-agent reinforcement learning, game-theoretic modeling, and adaptive matching mechanisms, it enables online, decentralized teaming decisions in non-stationary populations. Experiments demonstrate that our framework significantly outperforms fixed-group and unidirectional baselines across mainstream MARL tasks, while exhibiting superior policy generalization—maintaining robust collaborative performance under unseen population sizes and interaction topologies. This work establishes both theoretical foundations and practical methodologies for bidirectional collaborative learning in dynamic multi-agent systems.

Technology Category

• Multiagent Systems: Teamwork
• Humans and AI: Teamwork, Team formation
• Game Theory and Economic Paradigms: Coordination and Collaboration

Application Category

• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• 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

Team formation and the dynamics of team-based learning have drawn significant interest in the context of Multi-Agent Reinforcement Learning (MARL). However, existing studies primarily focus on unilateral groupings, predefined teams, or fixed-population settings, leaving the effects of algorithmic bilateral grouping choices in dynamic populations underexplored. To address this gap, we introduce a framework for learning two-sided team formation in dynamic multi-agent systems. Through this study, we gain insight into what algorithmic properties in bilateral team formation influence policy performance and generalization. We validate our approach using widely adopted multi-agent scenarios, demonstrating competitive performance and improved generalization in most scenarios.

Problem

• Explores bilateral team formation in dynamic multi-agent systems
• Investigates algorithmic properties affecting policy performance and generalization
• Addresses gaps in dynamic population team formation studies

Innovation

• Bilateral team formation in dynamic systems
• Algorithmic properties influence policy performance
• Improved generalization in multi-agent scenarios