Language-conditioned reinforcement learning to solve misunderstandings with action corrections

Röder, Frank; Eppe, Manfred

Language-conditioned reinforcement learning to solve misunderstandings with action corrections

Publikationstyp

Conference Paper not in Proceedings

Publikationsdatum

2022-11-17

Sprache

English

Author

Röder, Frank

Eppe, Manfred

Institut

Data Science Foundations E-21

TORE-URI

http://hdl.handle.net/11420/14361

Citation

36th Conference on Neural Information Processing Systems (NeurIPS 2022)

Contribution to Conference

36th Conference on Neural Information Processing Systems, NeurIPS 2022

ArXiv ID

2211.10168

Peer Reviewed

true

Cites

https://openreview.net/forum?id=lWd0qiv9E-

Human-to-human conversation is not just talking and listening. It is an incremental process where participants continually establish a common understanding to rule out misunderstandings. Current language understanding methods for intelligent robots do not consider this. There exist numerous approaches considering non-understandings, but they ignore the incremental process of resolving misunderstandings. In this article, we present a first formalization and experimental validation of incremental action-repair for robotic instruction-following based on reinforcement learning. To evaluate our approach, we propose a collection of benchmark environments for action correction in language-conditioned reinforcement learning, utilizing a synthetic instructor to generate language goals and their corresponding corrections. We show that a reinforcement learning agent can successfully learn to understand incremental corrections of misunderstood instructions.