On the role of population heterogeneity in emergent communication

Rita, Mathieu; Strub, Florian; Grill, Jean-Bastien; Pietquin, Olivier; Dupoux, Emmanuel

Document type :

Communication dans un congrès avec actes

Title :

On the role of population heterogeneity in emergent communication

Author(s) :

Rita, Mathieu [Auteur]
Apprentissage machine et développement cognitif [CoML]
Laboratoire de sciences cognitives et psycholinguistique [LSCP]
Strub, Florian [Auteur]
Sequential Learning [SEQUEL]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Grill, Jean-Bastien [Auteur]
DeepMind [Paris]
Pietquin, Olivier [Auteur]
Google Brain, Paris
Dupoux, Emmanuel [Auteur]
École des hautes études en sciences sociales [EHESS]

Apprentissage machine et développement cognitif [CoML]
Laboratoire de sciences cognitives et psycholinguistique [LSCP]

Conference title :

ICLP 2022 - Tenth International Conference on Learning Representations

City :

Los Angeles

Country :

Etats-Unis d'Amérique

Start date of the conference :

2022-04-25

HAL domain(s) :

Sciences cognitives/Informatique
Sciences cognitives/Linguistique

English abstract : [en]

Populations have often been perceived as a structuring component for language to emerge and evolve: the larger the population, the more structured the language. While this observation is widespread in the sociolinguistic ...
Show more >Populations have often been perceived as a structuring component for language to emerge and evolve: the larger the population, the more structured the language. While this observation is widespread in the sociolinguistic literature, it has not been consistently reproduced in computer simulations with neural agents. In this paper, we thus aim to clarify this apparent contradiction. We explore emergent language properties by varying agent population size in the speaker-listener Lewis Game. After reproducing the experimental difference, we challenge the simulation assumption that the agent community is homogeneous. We then investigate how speaker-listener asymmetry alters language structure through the analysis a potential diversity factor: learning speed. From then, we leverage this observation to control population heterogeneity without introducing confounding factors. We finally show that introducing such training speed heterogeneities naturally sort out the initial contradiction: larger simulated communities start developing more stable and structured languages.Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

ANR Project :

Frontières en cognition
PaRis Artificial Intelligence Research InstitutE

Collections :