EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Agent Coordination Through Emergent Symbolic Communication Under Bandwidth Limits

Tiffany Tsz-Kit Luk, Chris Yu

INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES · 23.09.2026 · 6 (2) · 165-177

DOI: 10.66238/ijrss.v6i2.136 ↗

Kurz gefasst

Fachartikel in der Zeitschrift INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES, online erschienen am 23.09.2026. Verfasst von Tiffany Tsz-Kit Luk und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The capacity for autonomous agents to coordinate their actions in complex environments is increasingly contingent upon their ability to communicate effectively. However, in many practical applications, such as distributed robotic swarms and remote sensor networks, communication channels are severely constrained by physical bandwidth limitations. This paper investigates the phenomenon of emergent symbolic communication among learning agents operating under strict bandwidth constraints. By framing the coordination problem within a multi-agent reinforcement learning paradigm, we explore how discrete, low-dimensional symbolic tokens can evolve to encapsulate high-dimensional environmental observations and strategic intentions. The study systematically evaluates the impact of varying channel capacities on the structural properties of the emergent lexicon and the consequent coordination performance of the agent collective. Through comprehensive theoretical analysis and simulated experimental environments, we demonstrate that optimal bandwidth restriction serves as a regularization mechanism, forcing agents to develop highly compositional and generalizable communication protocols. These emergent languages not only reduce the communication overhead but also exhibit structural similarities to natural language, specifically in terms of synonymy, polysemy, and context-dependent disambiguation. The findings provide critical insights into the design of efficient, robust multi-agent systems and contribute to the broader understanding of language evolution in artificial intelligence.

Zitieren

APA

Luk, T. T., & Yu, C. (2026). Agent Coordination Through Emergent Symbolic Communication Under Bandwidth Limits. INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES, 6(2), 165-177. https://doi.org/10.66238/ijrss.v6i2.136

Vancouver

Luk TT, Yu C. Agent Coordination Through Emergent Symbolic Communication Under Bandwidth Limits. INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES. 2026;6(2):165-177. doi:10.66238/ijrss.v6i2.136

BibTeX
@article{Luk2026,
  author = {Luk, Tiffany Tsz-Kit and Yu, Chris},
  title = {Agent Coordination Through Emergent Symbolic Communication Under Bandwidth Limits},
  journal = {INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES},
  year = {2026},
  volume = {6},
  number = {2},
  pages = {165--177},
  doi = {10.66238/ijrss.v6i2.136}
}
Kostenlos starten →