Compositional Languages Emerge in a Neural Iterated Learning Model
This addresses the problem of enabling efficient and generalizable communication in AI systems, though it is incremental as it builds on existing iterated learning models.
The paper tackles the emergence of compositional languages in neural agents through a neural iterated learning (NIL) algorithm, showing that these languages improve learning speed and generalization, with experiments confirming significant gains in agent performance.
The principle of compositionality, which enables natural language to represent complex concepts via a structured combination of simpler ones, allows us to convey an open-ended set of messages using a limited vocabulary. If compositionality is indeed a natural property of language, we may expect it to appear in communication protocols that are created by neural agents in language games. In this paper, we propose an effective neural iterated learning (NIL) algorithm that, when applied to interacting neural agents, facilitates the emergence of a more structured type of language. Indeed, these languages provide learning speed advantages to neural agents during training, which can be incrementally amplified via NIL. We provide a probabilistic model of NIL and an explanation of why the advantage of compositional language exist. Our experiments confirm our analysis, and also demonstrate that the emerged languages largely improve the generalizing power of the neural agent communication.