Paper: Reinforcement Learning for Mapping Instructions to Actions

ACL ID P09-1010
Title Reinforcement Learning for Mapping Instructions to Actions
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2009

In this paper, we present a reinforce- ment learning approach for mapping nat- ural language instructions to sequences of executable actions. We assume access to a reward function that defines the qual- ity of the executed actions. During train- ing, the learner repeatedly constructs ac- tion sequences for a set of documents, ex- ecutes those actions, and observes the re- sulting 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 troubleshoot- ing guides and game tutorials. Our results demonstrate that this method can rival su- pervised learning techniques while requir- ing few or no annotated training exam- ples.1