Situated Mapping of Sequential Instructions to Actions with Single-step Reward Observation

Alane Suhr, Yoav Artzi


Abstract
We propose a learning approach for mapping context-dependent sequential instructions to actions. We address the problem of discourse and state dependencies with an attention-based model that considers both the history of the interaction and the state of the world. To train from start and goal states without access to demonstrations, we propose SESTRA, a learning algorithm that takes advantage of single-step reward observations and immediate expected reward maximization. We evaluate on the SCONE domains, and show absolute accuracy improvements of 9.8%-25.3% across the domains over approaches that use high-level logical representations.
Anthology ID:
P18-1193
Volume:
Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2018
Address:
Melbourne, Australia
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2072–2082
Language:
URL:
https://aclanthology.org/P18-1193
DOI:
10.18653/v1/P18-1193
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/P18-1193.pdf
Note:
 P18-1193.Notes.pdf
Video:
 https://vimeo.com/285805263
Presentation:
 P18-1193.Presentation.pdf
Code
 clic-lab/scone
Data
ATIS