Emergent Communication in Continuous Worlds: Self-Organisation of Conceptually Grounded Vocabularies at Scale

Jérôme Botoko Ekila, Lara Verheyen, Jens Nevens, Katrien Beuls, Paul Van Eecke

Keywords: emergent communication, multi-agent systems, language games, language emergence, language evolution

TL;DR: Methodology for the de-centralised emergence of conceptually grounded linguistic conventions in populations of agents, widely applicable to continuous and categorical datasets.

Abstract: This paper introduces a general methodology through which a population of autonomous agents can converge on a linguistic convention that enables them to refer to arbitrary entities in their environment. The linguistic convention emerges in a decentralised manner through local communicative interactions between pairs of agents drawn from the population. The emergent convention consists of associations between symbolic labels (word forms) and subsymbolic concept representations (word meanings) that are grounded in a continuous feature space. We confirm the generality and scalability of the method through its evaluation on a wide and diverse selection of 34 publicly available datasets. We also demonstrate the robustness of the method against perceptual variation, including in heteromorphic populations, as well as the ability of the emergent conventions to self-adapt to changes in the environment.

Submission Number: 58