On the Spontaneous Emergence of Discrete and Compositional Signals

Nur Geffen Lan, Emmanuel Chemla, Shane Steinert-Threlkeld


Abstract
We propose a general framework to study language emergence through signaling games with neural agents. Using a continuous latent space, we are able to (i) train using backpropagation, (ii) show that discrete messages nonetheless naturally emerge. We explore whether categorical perception effects follow and show that the messages are not compositional.
Anthology ID:
2020.acl-main.433
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Editors:
Dan Jurafsky, Joyce Chai, Natalie Schluter, Joel Tetreault
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4794–4800
Language:
URL:
https://aclanthology.org/2020.acl-main.433
DOI:
10.18653/v1/2020.acl-main.433
Bibkey:
Cite (ACL):
Nur Geffen Lan, Emmanuel Chemla, and Shane Steinert-Threlkeld. 2020. On the Spontaneous Emergence of Discrete and Compositional Signals. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4794–4800, Online. Association for Computational Linguistics.
Cite (Informal):
On the Spontaneous Emergence of Discrete and Compositional Signals (Geffen Lan et al., ACL 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.acl-main.433.pdf
Video:
 http://slideslive.com/38929275
Code
 0xnurl/signaling-auto-encoder