Abstract:Constructive studies on symbol emergence systems seek to investigate computational models that can better explain human language evolution, the creation of symbol systems, and the construction of internal representations. Specifically, emergent communication aims to formulate a computational model that enables agents to build efficient sign systems and internal representations. This study provides a new model for emergent communication, which is based on a probabilistic generative model (PGM) instead of a discriminative model based on deep reinforcement learning. We define the Metropolis-Hastings (MH) naming game by generalizing previously proposed models. It is not a referential game with explicit feedback, as assumed by many emergent communication studies. Instead, it is a game based on joint attention without explicit feedback. Mathematically, the MH naming game is proved to be a type of MH algorithm for an integrative PGM that combines two agents that play the naming game. From this viewpoint, symbol emergence is regarded as decentralized Bayesian inference , and semiotic communication is regarded as inter-personal cross-modal inference. This notion leads to the collective predictive coding hypothesis regarding language evolution and, in general, the emergence of symbols. We also propose the inter-Gaussian mixture model (GMM)+ variational autoencoder (VAE), a deep generative model for emergent communication based on the MH naming game. In this model, two agents create internal representations and categories and share signs (i.e. names of objects) from raw visual images observed from different viewpoints. The model has been validated on MNIST and Fruits 360 datasets. Experimental findings demonstrate that categories are formed from real images observed by agents, and signs are correctly shared across agents by successfully utilizing both of the observations of agents via the MH naming game. Furthermore, scholars verified that visual images were recalled from signs uttered by agents. Notably, emergent communication without supervision and reward feedback improved the performance of the unsupervised representation learning of agents.

Emergent Communication for Numerical Concepts Generalization

Emergence of Numeric Concepts in Multi-Agent Autonomous Communication

Emergent Communication of Generalizations

Emergent Communication for Rules Reasoning

The Emergence of Compositional Languages for Numeric Concepts Through Iterated Learning in Neural Agents

Learning Multi-Object Positional Relationships Via Emergent Communication.

Emergent communication through Metropolis-Hastings naming game with deep generative models

The Challenge of Modeling the Acquisition of Mathematical Concepts

A Combinatorial Approach to Neural Emergent Communication

Emergent Graphical Conventions in a Visual Communication Game

Learning Approximate and Exact Numeral Systems via Reinforcement Learning

Emergence of Linguistic Communication from Referential Games with Symbolic and Pixel Input

On Emergent Communication in Competitive Multi-Agent Teams

Incorporating Pragmatic Reasoning Communication into Emergent Language

Expressivity of Emergent Language is a Trade-off between Contextual Complexity and Unpredictability

Compositionality and Generalization in Emergent Languages

MathKingdom: Teaching Children Mathematical Language Through Speaking at Home Via a Voice-Guided Game.

Emergent Communication: Generalization and Overfitting in Lewis Games

EGG: a toolkit for research on Emergence of lanGuage in Games

Game Description Logic with Integers: A GDL Numerical Extension

ESCELL: Emergent Symbolic Cellular Language