Decentralized Multi-Agent World Models and Emergent Communication

2024年1月1日 · 1 min read
research

Building on the framework of collective predictive coding, we computationally model how multiple agents can learn world models in a decentralized way while emerging communication symbols through interaction. We clarify the conditions under which collective world models function as the basis for linguistic transmission and coordinated behavior.

Related papers:

  • Kentaro Nomura, Tatsuya Aoki, Tadahiro Taniguchi, Takato Horii. “Decentralized Collective World Model for Emergent Communication and Coordination”. 2025 IEEE International Conference on Development and Learning (ICDL), September 2025.
  • Kentaro Nomura, Tatsuya Aoki, Tadahiro Taniguchi, Takato Horii. “Decentralized Multi-Agent World Model Trainable via Collective Predictive Coding” (in Japanese). Annual Conference of the Japanese Society for Artificial Intelligence (JSAI), May 2025.
Social-Reality-Model
Takato Horii
Authors
Takato Horii (he/him)
Associate Professor
Associate Professor at Graduate School of Engineering Science, Osaka University. Research interests include cognitive developmental robotics, computational modeling of emotional development, and human-robot interaction.