Imitation learning for language generation from unaligned data.

Natural language generation (NLG) is the task of generating natural language from a meaning representation. Rule-based approaches require domain-specific and manually constructed linguistic resources, while most corpus based approaches rely on aligned training data and/or phrase templates. The latter are needed to restrict the search space for the structured prediction task defined by the unaligned datasets. In this work we propose the use of imitation learning for structured prediction which learns an incremental model that handles the large search space while avoiding explicitly enumerating it. We adapted the Locally Optimal Learning to Search (Chang et al., 2015) framework which allows us to train against non-decomposable loss functions such as the BLEU or ROUGE scores while not assuming gold standard alignments. We evaluate our approach on three datasets using both automatic measures and human judgements and achieve results comparable to the state-of-the-art approaches developed for each of them. Furthermore, we performed an analysis of the datasets which examines common issues with NLG evaluation.