Communicating artificial neural networks develop efficient color-naming systems release_evprhe4zjbacna2v7ws47seuz4

by Rahma Chaabouni, Eugene Kharitonov, Emmanuel Dupoux, Marco Baroni

Published in Proceedings of the National Academy of Sciences of the United States of America by Proceedings of the National Academy of Sciences.

2021   Volume 118, Issue 12, e2016569118

Abstract

Words categorize the semantic fields they refer to in ways that maximize communication accuracy while minimizing complexity. Focusing on the well-studied color domain, we show that artificial neural networks trained with deep-learning techniques to play a discrimination game develop communication systems whose distribution on the accuracy/complexity plane closely matches that of human languages. The observed variation among emergent color-naming systems is explained by different degrees of discriminative need, of the sort that might also characterize different human communities. Like human languages, emergent systems show a preference for relatively low-complexity solutions, even at the cost of imperfect communication. We demonstrate next that the nature of the emergent systems crucially depends on communication being discrete (as is human word usage). When continuous message passing is allowed, emergent systems become more complex and eventually less efficient. Our study suggests that efficient semantic categorization is a general property of discrete communication systems, not limited to human language. It suggests moreover that it is exactly the discrete nature of such systems that, acting as a bottleneck, pushes them toward low complexity and optimal efficiency.
In application/xml+jats format

Archived Files and Locations

application/pdf  1.1 MB
file_qnv4bdlmk5arhivs4ygf3xp3c4
www.pnas.org (publisher)
web.archive.org (webarchive)
application/pdf  1.1 MB
file_bnhpmwkrgzbndpe3red3wgdi4m
repositori.upf.edu (web)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article-journal
Stage   published
Date   2021-03-15
Language   en ?
Proceedings Metadata
Open Access Publication
Not in DOAJ
In Keepers Registry
ISSN-L:  0027-8424
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: cf76675b-c7ec-470e-b685-3d6d7a0d34a3
API URL: JSON