A Combinatorial Approach to Neural Emergent Communication

TL;DR

This paper introduces a combinatorial algorithm to analyze and enhance emergent communication in neural networks, addressing limitations of existing referential game frameworks by increasing symbolic complexity.

Contribution

It provides a theoretical analysis and a novel algorithm, SolveMinSym, to measure and improve symbolic complexity in emergent communication datasets.

Findings

Higher symbolic complexity leads to more effective symbols in emergent language

The SMS algorithm successfully creates datasets with controlled symbolic complexity

Increased complexity improves communication effectiveness in neural models

Abstract

Substantial research on deep learning-based emergent communication uses the referential game framework, specifically the Lewis signaling game, however we argue that successful communication in this game typically only need one or two symbols for target image classification because of a sampling pitfall in the training data. To address this issue, we provide a theoretical analysis and introduce a combinatorial algorithm SolveMinSym (SMS) to solve the symbolic complexity for classification, which is the minimum number of symbols in the message for successful communication. We use the SMS algorithm to create datasets with different symbolic complexity to empirically show that data with higher symbolic complexity increases the number of effective symbols in the emergent language.