Discovering Properties of Inflectional Morphology in Neural Emergent Communication
By: Miles Gilberti, Shane Storks, Huteng Dai
Potential Business Impact:
Computers learn to make new words like humans.
Emergent communication (EmCom) with deep neural network-based agents promises to yield insights into the nature of human language, but remains focused primarily on a few subfield-specific goals and metrics that prioritize communication schemes which represent attributes with unique characters one-to-one and compose them syntactically. We thus reinterpret a common EmCom setting, the attribute-value reconstruction game, by imposing a small-vocabulary constraint to simulate double articulation, and formulating a novel setting analogous to naturalistic inflectional morphology (enabling meaningful comparison to natural language communication schemes). We develop new metrics and explore variations of this game motivated by real properties of inflectional morphology: concatenativity and fusionality. Through our experiments, we discover that simulated phonological constraints encourage concatenative morphology, and emergent languages replicate the tendency of natural languages to fuse grammatical attributes.
Similar Papers
Walk the Robot: Exploring Soft Robotic Morphological Communication driven by Spiking Neural Networks
Neural and Evolutionary Computing
Robot parts talk to each other using wiggles.
Transformer-Enabled Diachronic Analysis of Vedic Sanskrit: Neural Methods for Quantifying Types of Language Change
Computation and Language
Uncovers how ancient languages change over time.
Why are LLMs' abilities emergent?
Computation and Language
AI learns new skills by itself, like magic.