Decentralized Collective World Model for Emergent Communication and Coordination
This addresses the challenge of integrating communication and coordination in decentralized multi-agent systems, representing an incremental advance over prior work that treated these aspects separately.
The paper tackles the problem of enabling both communication and coordination in multi-agent systems through a decentralized world model, demonstrating that their approach outperforms non-communicative models in a trajectory drawing task and facilitates meaningful symbol emergence.
We propose a fully decentralized multi-agent world model that enables both symbol emergence for communication and coordinated behavior through temporal extension of collective predictive coding. Unlike previous research that focuses on either communication or coordination separately, our approach achieves both simultaneously. Our method integrates world models with communication channels, enabling agents to predict environmental dynamics, estimate states from partial observations, and share critical information through bidirectional message exchange with contrastive learning for message alignment. Using a two-agent trajectory drawing task, we demonstrate that our communication-based approach outperforms non-communicative models when agents have divergent perceptual capabilities, achieving the second-best coordination after centralized models. Importantly, our decentralized approach with constraints preventing direct access to other agents' internal states facilitates the emergence of more meaningful symbol systems that accurately reflect environmental states. These findings demonstrate the effectiveness of decentralized communication for supporting coordination while developing shared representations of the environment.