Dynamics of Language and Cognition Based on Collective Predictive Coding: Toward New Developments in Symbolic Emergent Robotics.
This paper introduces the new concept of “collective predictive coding” and proposes a “collective predictive coding hypothesis” that explains language emergence and evolution. The main contents of the paper are as follows:
- 1.Symbolic emergent system: A model that explains the process by which symbolic systems, including language, emerge through individual cognition and social interaction.
- 2.collective predictive coding: A process in which multiple agents collectively learn a model of the world by performing Bayesian inference in a distributed manner.
- 3.Collective Predictive Coding Hypothesis: The hypothesis that human language is formed by collective predictive coding.
- new developments in symbolic emergent robotics: Based on this hypothesis, we suggest the possibility of developing more complex models and co-creative learning between humans and robots.
- relationship with existing approaches: Discuss the relationship with existing research such as emergent communication, multi-agent reinforcement learning, iterative learning models, symbolic emergent robotics, etc.
- future work: extensions to the theory, experimental validation, and connections to computational linguistics.
This theory is an ambitious attempt to integrate individual cognition and social interaction into a comprehensive explanation of language emergence and evolution, and may offer new perspectives in cognitive science, linguistics, and artificial intelligence.
This page is auto-translated from /nishio/集合的予測符号化に基づく言語と認知のダイナミクス：記号創発ロボティクスの新展開に向けて using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.