Intrinsically Motivated Compositional Language Emergence
This work is significant for researchers in emergent communication, as it offers a new approach to developing more human-like compositional languages in artificial agents.
The paper addresses the lack of compositionality in emergent languages among artificial agents by proposing an intrinsic reward framework. This method improves compositionality scores by approximately 1.5-2 times compared to existing frameworks that rely solely on limited channel capacity.
Recently, there has been a great deal of research in emergent communication on artificial agents interacting in simulated environments. Recent studies have revealed that, in general, emergent languages do not follow the compositionality patterns of natural language. To deal with this, existing works have proposed a limited channel capacity as an important constraint for learning highly compositional languages. In this paper, we show that this is not a sufficient condition and propose an intrinsic reward framework for improving compositionality in emergent communication. We use a reinforcement learning setting with two agents -- a \textit{task-aware} Speaker and a \textit{state-aware} Listener that are required to communicate to perform a set of tasks. Through our experiments on three different referential game setups, including a novel environment gComm, we show intrinsic rewards improve compositionality scores by $\approx \mathbf{1.5-2}$ times that of existing frameworks that use limited channel capacity.