[TOC]

  1. Title: BABYAI++: Towards Grounded-Language Learning Beyond Memorization
  2. Author: Tianshi Cao et. al.
  3. Publish Year: 2020 ICLR
  4. Review Date: Jan 2022

Summary of paper

The paper introduced a new RL environment BabyAI++ that can investigate whether RL agents can extract knowledge from descriptive text and eventually increase generalisation performance.

BabyAI++ environment example

image-20220104181838965

Model

they studied three architecture for processing descriptive text and scene representation into a combined visual+text embedding

  1. concat-fusion
  2. FiLM
  3. attention fusion

their attention fusion model obtained the best result

image-20220104201742244

Contribution

Their results shows that attention based text-image grounded language learning allows the agents to extract knowledge from descriptive text and obtain better generalisation performance.