Discovering the Universality of Language through Deep Learning

Published on June 4, 2022

In this fascinating paper, researchers use deep learning to explore the concept of non-arbitrariness in language. Using different sequence-processing neural networks, they assess the prevalence of non-arbitrary phonological patterns in various languages. Through a series of experiments, they train the models to associate the sounds of words with their sensory, semantic, and word-class representations. The exciting part is that these trained models are then tested on completely new instances in languages from different language families – and they perform remarkably well! This suggests that there are universal cues in the phonological structure of words that provide insights into their meaning and syntax, regardless of the language they belong to. This study not only sheds light on the universality of language but also demonstrates the power of deep learning in uncovering hidden linguistic patterns. For more details on their methodology and findings, check out the full article!

Abstract
The present paper addresses the study of non-arbitrariness in language within a deep learning framework. We present a set of experiments aimed at assessing the pervasiveness of different forms of non-arbitrary phonological patterns across a set of typologically distant languages. Different sequence-processing neural networks are trained in a set of languages to associate the phonetic vectorization of a set of words to their sensory (Experiment 1), semantic (Experiment 2), and word-class representations (Experiment 3). The models are then tested, without further training, in a set of novel instances in a language belonging to a different language family, and their performance is compared with a randomized baseline. We show that the three cross-domain mappings can be successfully transferred across languages and language families, suggesting that the phonological structure of the lexicon is pervaded with language-invariant cues about the words’ meaning and their syntactic classes.

Read Full Article (External Site)

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>