Emergence of Compositional Language with Deep Generational Transmission
This work addresses the challenge of developing more human-like, interpretable communication systems for AI agents, though it is incremental by building on prior studies of structural priors.
The paper tackled the problem of encouraging compositional language emergence in deep reinforcement learning agents by introducing cultural evolutionary dynamics through periodic agent replacement, resulting in languages that exhibit better compositional generalization.
Recent work has studied the emergence of language among deep reinforcement learning agents that must collaborate to solve a task. Of particular interest are the factors that cause language to be compositional -- i.e., express meaning by combining words which themselves have meaning. Evolutionary linguists have found that in addition to structural priors like those already studied in deep learning, the dynamics of transmitting language from generation to generation contribute significantly to the emergence of compositionality. In this paper, we introduce these cultural evolutionary dynamics into language emergence by periodically replacing agents in a population to create a knowledge gap, implicitly inducing cultural transmission of language. We show that this implicit cultural transmission encourages the resulting languages to exhibit better compositional generalization.