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 txt2π, a model that captures three-way interactions between the goal, document, and observations. On RTFM, txt2π 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, txt2π produces policies that excel on complex RTFM tasks requiring several reasoning and coreference steps.
RTFM: Generalising to New Environment Dynamics via Reading
2020-01-01
In: Proceedings of the International Conference on Learning Representations (ICLR 2020). (pp. pp. 1-17). ICLR (2020)
Paper
Electronic Resource
English
DDC: | 629 |
Generalising results for any legal initial configuration of a railway network
Tema Archive | 1996
|British Library Conference Proceedings | 2001
|Thoughts on Reading Artigue's ``Learning Mathematics in a CAS Environment''
British Library Online Contents | 2002
|Reading Municipal Airport, Reading, Pennsylvania
NTIS | 1971