Emergence of Compositional Language with Deep Generational Transmission
Michael Cogswell, Jiasen Lu, Stefan Lee, Devi Parikh, Dhruv Batra

TL;DR
This paper demonstrates that introducing cultural transmission dynamics in deep reinforcement learning agents enhances the emergence of compositional language, leading to better generalization of communication.
Contribution
It introduces a novel method of simulating cultural transmission through periodic agent replacement to promote compositional language emergence.
Findings
Cultural transmission improves compositional generalization.
Periodic agent replacement fosters language structure.
Languages exhibit better generalization with transmission dynamics.
Abstract
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.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsLanguage and cultural evolution · Origins and Evolution of Life · Evolutionary Game Theory and Cooperation
