Reinforcement Learning for Mapping Instructions to Actions
Name
Barzilay_Reinforcement learning.pdf
Size
610.8 KB
Format
Adobe PDF
Checksum (MD5)
20ec0421e2dce6276e3aba4dfdd04311
Author(s)
Branavan, Satchuthanan R.
Chen, Harr
Zettlemoyer, Luke S.
Barzilay, Regina
Date Issued
August 2009
Journal
Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP
Publisher
Association for Computational Linguistics
Citation
Branavan, S.R.K., Harr Chen, Luke S. Zettlemoyer, and Regina Barzilay (2009). "Reinforcement learning for mapping instructions to actions." Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP (Morristown, N.J.: Association for Computational Linguistics): 82-90. © Association for Computing Machinery.
Version
Author's final manuscript
Abstract
In this paper, we present a reinforcement learning approach for mapping natural language instructions to sequences of executable actions. We assume access to a reward function that defines the quality of the executed actions. During training, the learner repeatedly constructs action sequences for a set of documents, executes those actions, and observes the resulting reward. We use a policy gradient algorithm to estimate the parameters of a log-linear model for action selection. We apply our method to interpret instructions in two domains --- Windows troubleshooting guides and game tutorials. Our results demonstrate that this method can rival supervised learning techniques while requiring few or no annotated training examples.
Subjects
algorithms
design
experimentation
languages
measurement
performance
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Attribution-Noncommercial-Share Alike 3.0 Unported
Persistent DSpace Link
DOI of Published Version
http://portal.acm.org/citation.cfm?id=1687892
