Inferring Rewards from Language in Context

Jessy Lin, Daniel Fried, Dan Klein, Anca Dragan


Abstract
In classic instruction following, language like “I’d like the JetBlue flight” maps to actions (e.g., selecting that flight). However, language also conveys information about a user’s underlying reward function (e.g., a general preference for JetBlue), which can allow a model to carry out desirable actions in new contexts. We present a model that infers rewards from language pragmatically: reasoning about how speakers choose utterances not only to elicit desired actions, but also to reveal information about their preferences. On a new interactive flight–booking task with natural language, our model more accurately infers rewards and predicts optimal actions in unseen environments, in comparison to past work that first maps language to actions (instruction following) and then maps actions to rewards (inverse reinforcement learning).
Anthology ID:
2022.acl-long.585
Volume:
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
May
Year:
2022
Address:
Dublin, Ireland
Editors:
Smaranda Muresan, Preslav Nakov, Aline Villavicencio
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8546–8560
Language:
URL:
https://aclanthology.org/2022.acl-long.585
DOI:
10.18653/v1/2022.acl-long.585
Bibkey:
Cite (ACL):
Jessy Lin, Daniel Fried, Dan Klein, and Anca Dragan. 2022. Inferring Rewards from Language in Context. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8546–8560, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Inferring Rewards from Language in Context (Lin et al., ACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.acl-long.585.pdf
Video:
 https://aclanthology.org/2022.acl-long.585.mp4
Code
 jlin816/rewards-from-language