Inductive Bias and Language Expressivity in Emergent Communication
This research is significant for understanding the properties of emergent languages in artificial intelligence systems, particularly for researchers working on multi-agent communication.
This paper investigates how the type of language game (referential vs. reconstruction) influences the emergent language's compositionality and expressivity. The authors demonstrate empirically on a symbolic dataset that different game types lead to languages with distinct compositionality and expressivity.
Referential games and reconstruction games are the most common game types for studying emergent languages. We investigate how the type of the language game affects the emergent language in terms of: i) language compositionality and ii) transfer of an emergent language to a task different from its origin, which we refer to as language expressivity. With empirical experiments on a handcrafted symbolic dataset, we show that languages emerged from different games have different compositionality and further different expressivity.