BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learning

TL;DR

The paper introduces the BabyAI platform to study how efficiently artificial agents can learn grounded language instructions through human interaction, highlighting current deep learning limitations in sample efficiency for compositional language learning.

Contribution

It presents a new platform with diverse levels and a heuristic expert to facilitate research on language learning efficiency and human-in-the-loop training methods.

Findings

Current deep learning methods lack sufficient sample efficiency for compositional language learning.

Baseline results demonstrate the amount of human involvement needed for training agents.

BabyAI provides a structured environment for systematic investigation of grounded language learning.

Abstract

Allowing humans to interactively train artificial agents to understand language instructions is desirable for both practical and scientific reasons, but given the poor data efficiency of the current learning methods, this goal may require substantial research efforts. Here, we introduce the BabyAI research platform to support investigations towards including humans in the loop for grounded language learning. The BabyAI platform comprises an extensible suite of 19 levels of increasing difficulty. The levels gradually lead the agent towards acquiring a combinatorially rich synthetic language which is a proper subset of English. The platform also provides a heuristic expert agent for the purpose of simulating a human teacher. We report baseline results and estimate the amount of human involvement that would be required to train a neural network-based agent on some of the BabyAI levels. We put forward strong evidence that current deep learning methods are not yet sufficiently sample efficient when it comes to learning a language with compositional properties.