Low-Rank Tensors for Scoring Dependency Structures
Accurate scoring of syntactic structures such as head-modifier arcs in dependency parsing typically requires rich, high-dimensional feature representations. A small subset of such features is often selected manually. This is problematic when features lack clear linguistic meaning as in embeddings o...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Association for Computational Linguistics
2015
|
| Online Access: | http://hdl.handle.net/1721.1/99745 https://orcid.org/0000-0003-3121-0185 https://orcid.org/0000-0002-2921-8201 https://orcid.org/0000-0002-2199-0379 https://orcid.org/0000-0003-4644-3088 |
| Summary: | Accurate scoring of syntactic structures such as head-modifier arcs in dependency parsing typically requires rich, high-dimensional feature representations. A small subset of such features is often selected manually. This is problematic when features lack clear linguistic meaning as in embeddings or when the information is blended across features. In this paper, we use tensors to map high-dimensional feature vectors into low dimensional representations. We explicitly maintain the parameters as a low-rank tensor to obtain low dimensional representations of words in their syntactic roles, and to leverage modularity in the tensor for easy training with online algorithms. Our parser consistently outperforms the Turbo and MST parsers across 14 different languages. We also obtain the best published UAS results on 5 languages. |
|---|