Emergent Multi-Agent Communication in the Deep Learning Era
It addresses the problem of enabling AI systems to communicate effectively for both scientific insights into human language evolution and practical applications in everyday life, but is incremental as it is a survey.
The paper surveys recent studies on whether deep artificial networks can develop a shared language for cooperation, exploring conditions for language evolution and its potential to enhance network flexibility in solving problems interactively.
The ability to cooperate through language is a defining feature of humans. As the perceptual, motory and planning capabilities of deep artificial networks increase, researchers are studying whether they also can develop a shared language to interact. From a scientific perspective, understanding the conditions under which language evolves in communities of deep agents and its emergent features can shed light on human language evolution. From an applied perspective, endowing deep networks with the ability to solve problems interactively by communicating with each other and with us should make them more flexible and useful in everyday life. This article surveys representative recent language emergence studies from both of these two angles.