Structured Attention Networks

@article{Kim2016StructuredAN,
  title={Structured Attention Networks},
  author={Yoon Kim and Carl Denton and Luong Hoang and Alexander M. Rush},
  journal={CoRR},
  year={2016},
  volume={abs/1702.00887}
}
Attention networks have proven to be an effective approach for embedding categorical inference within a deep neural network. However, for many tasks we may want to model richer structural dependencies without abandoning end-to-end training. In this work, we experiment with incorporating richer structural distributions, encoded using graphical models, within deep networks. We show that these structured attention networks are simple extensions of the basic attention procedure, and that they allow… CONTINUE READING
Highly Cited
This paper has 86 citations. REVIEW CITATIONS
Recent Discussions
This paper has been referenced on Twitter 127 times over the past 90 days. VIEW TWEETS
Blog posts, news articles and tweet counts and IDs sourced by
Altmetric.com