Paper: Constituent Parsing with Incremental Sigmoid Belief Networks

ACL ID P07-1080
Title Constituent Parsing with Incremental Sigmoid Belief Networks
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2007
Authors

We introduce a framework for syntactic parsing with latent variables based on a form of dynamic Sigmoid Belief Networks called Incremental Sigmoid Belief Networks. We demonstrate that a previous feed-forward neural network parsing model can be viewed as a coarse approximation to inference with this class of graphical model. By construct- ing a more accurate but still tractable ap- proximation, we significantly improve pars- ing accuracy, suggesting that ISBNs provide a good idealization for parsing. This gener- ative model of parsing achieves state-of-the- art results on WSJ text and 8% error reduc- tion over the baseline neural network parser.