RTFM: Generalising to Novel Environment Dynamics via Reading

Abstract

Obtaining policies that can generalise to new environments in reinforcement learning is challenging. In this work, we demonstrate that language understanding via a reading policy learner is a promising vehicle for generalisation to new environments. We propose a grounded policy learning problem, Read to Fight Monsters (RTFM), in which the agent must jointly reason over a language goal, relevant dynamics described in a document, and environment observations. We procedurally generate environment dynamics and corresponding language descriptions of the dynamics, such that agents must read to understand new environment dynamics instead of memorising any particular information. In addition, we propose txt2$\pi$, a model that captures three-way interactions between the goal, document, and observations. On RTFM, txt2$\pi$ generalises to new environments with dynamics not seen during training via reading. Furthermore, our model outperforms baselines such as FiLM and language-conditioned CNNs on RTFM. Through curriculum learning, txt2$\pi$ produces policies that excel on complex RTFM tasks requiring several reasoning and coreference steps.

arXiv commentICLR 2020; 17 pages, 13 figures

References in corpus (7)

• Continuous control with deep reinforcement learning
• Playing Atari with Deep Reinforcement Learning
• IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
• Deep Reinforcement Learning for Dialogue Generation
• Grounded Language Learning in a Simulated 3D World
• Global-Locally Self-Attentive Dialogue State Tracker
• TorchBeast: A PyTorch Platform for Distributed RL

Cited by in corpus (4)

• A Survey of Zero-shot Generalisation in Deep Reinforcement Learning
• Human Instruction-Following with Deep Reinforcement Learning via Transfer-Learning from Text
• ALFWorld: Aligning Text and Embodied Environments for Interactive Learning
• SILG: The Multi-environment Symbolic Interactive Language Grounding Benchmark