Symbol Emergence in Robotics: From Cognition to Language

This book presents a pioneering exploration of symbol and language emergence in cognitive systems. This novel research field, primarily developed by Japanese research groups, investigates the relationship between embodied cognition and language, centered around the concept of symbol emergence systems.

In the era of Large Language Models, understanding the roots and underlying mechanisms of language emergence is gaining attention. This book provides fresh insights by surveying cognitive developmental robotics and symbol emergence in robotics, focusing on language acquisition, concept formation, and symbol/language emergence in multi-agent systems.

The work addresses the symbol grounding problem in AI and cognitive science, arguing that it is an artificial construct. Instead, it posits the symbol emergence problem as a fundamentally important problem. Following this perspective, the book introduces constructive studies on robots acquiring language, covering developmental robotics, unsupervised multimodal category formation, and lexical acquisition in robots.

The "Collective Predictive Coding Hypothesis" is also introduced as a unified theory explaining how language and symbols emerge from cognitive processes in both humans and artificial systems. This hypothesis frames symbol emergence as decentralized Bayesian inference and explores a society-wide free energy principle.

The book also discusses the future of human-AI symbiotic society from the viewpoint of symbol emergence systems and the collective predictive coding hypothesis. It offers crucial insights for scientists, researchers, and professionals in robotics, AI, and cognitive science, providing a comprehensive understanding of symbol emergence and language acquisition in embodied systems.

This work is essential for anyone interested in the intersection of language and cognition in humans, artificial intelligence, and robots, offering a new perspective on how we might approach the development of more adaptable and context-aware AI systems and a human-AI symbiotic society in the future.