The Emergence of Compositional Languages for Numeric Concepts Through Iterated Learning in Neural Agents
This work addresses the problem of understanding language evolution in AI systems for researchers in computational linguistics and AI, but it is incremental as it builds on existing iterated learning frameworks.
The study tackled the emergence of compositional languages for numeric concepts in multi-agent systems, demonstrating that such languages can arise through iterated learning in neural agents, with emergence depending on input representations and requiring fewer learning iterations than non-degenerate alternatives.
Since first introduced, computer simulation has been an increasingly important tool in evolutionary linguistics. Recently, with the development of deep learning techniques, research in grounded language learning has also started to focus on facilitating the emergence of compositional languages without pre-defined elementary linguistic knowledge. In this work, we explore the emergence of compositional languages for numeric concepts in multi-agent communication systems. We demonstrate that compositional language for encoding numeric concepts can emerge through iterated learning in populations of deep neural network agents. However, language properties greatly depend on the input representations given to agents. We found that compositional languages only emerge if they require less iterations to be fully learnt than other non-degenerate languages for agents on a given input representation.