Lewis's Signaling Game as beta-VAE For Natural Word Lengths and Segments

TL;DR

This paper reinterprets Lewis's signaling game as a beta-VAE, showing how prior distributions influence emergent language properties like word length and segmentation, aligning them more closely with natural language laws.

Contribution

It reformulates the signaling game as a beta-VAE, clarifies the role of priors, and demonstrates how choosing appropriate priors leads to emergent languages with natural statistical properties.

Findings

Appropriate priors lead to emergent languages following Zipf's law of abbreviation.

Proper priors result in segments that align with Harris's articulation scheme.

Conventional objectives hinder emergent languages from exhibiting natural language properties.

Abstract

As a sub-discipline of evolutionary and computational linguistics, emergent communication (EC) studies communication protocols, called emergent languages, arising in simulations where agents communicate. A key goal of EC is to give rise to languages that share statistical properties with natural languages. In this paper, we reinterpret Lewis's signaling game, a frequently used setting in EC, as beta-VAE and reformulate its objective function as ELBO. Consequently, we clarify the existence of prior distributions of emergent languages and show that the choice of the priors can influence their statistical properties. Specifically, we address the properties of word lengths and segmentation, known as Zipf's law of abbreviation (ZLA) and Harris's articulation scheme (HAS), respectively. It has been reported that the emergent languages do not follow them when using the conventional objective. We experimentally demonstrate that by selecting an appropriate prior distribution, more natural segments emerge, while suggesting that the conventional one prevents the languages from following ZLA and HAS.