Learning to Draw: Emergent Communication through Sketching

TL;DR

This paper demonstrates that neural agents can learn to communicate through drawing in a differentiable framework, enabling interpretable visual communication that mimics early human prehistory and offers a flexible alternative to language-based methods.

Contribution

It introduces a differentiable drawing-based communication channel for neural agents, enabling end-to-end training and human-interpretable visual communication in collaborative tasks.

Findings

Agents successfully learn to communicate via drawing.

With proper biases, agents produce human-interpretable sketches.

Visual communication can be a flexible alternative to language in AI collaboration.

Abstract

Evidence that visual communication preceded written language and provided a basis for it goes back to prehistory, in forms such as cave and rock paintings depicting traces of our distant ancestors. Emergent communication research has sought to explore how agents can learn to communicate in order to collaboratively solve tasks. Existing research has focused on language, with a learned communication channel transmitting sequences of discrete tokens between the agents. In this work, we explore a visual communication channel between agents that are allowed to draw with simple strokes. Our agents are parameterised by deep neural networks, and the drawing procedure is differentiable, allowing for end-to-end training. In the framework of a referential communication game, we demonstrate that agents can not only successfully learn to communicate by drawing, but with appropriate inductive biases, can do so in a fashion that humans can interpret. We hope to encourage future research to consider visual communication as a more flexible and directly interpretable alternative of training collaborative agents.