Header logo is

Loss Functions and Optimization Methods for Discriminative Learning of Label Sequences

2003

Conference Paper

ei


Discriminative models have been of interest in the NLP community in recent years. Previous research has shown that they are advantageous over generative models. In this paper, we investigate how different objective functions and optimization methods affect the performance of the classifiers in the discriminative learning framework. We focus on the sequence labelling problem, particularly POS tagging and NER tasks. Our experiments show that changing the objective function is not as effective as changing the features included in the model.

Author(s): Altun, Y. and Johnson, M. and Hofmann, T.
Pages: 145-152
Year: 2003
Month: July
Day: 0
Editors: Collins, M. , M. Steedman
Publisher: ACL

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

Event Name: Conference on Empirical Methods in Natural Language Processing (EMNLP 2003)
Event Place: Sapporo, Japan

Address: East Stroudsburg, PA, USA
Digital: 0

Links: Web

BibTex

@inproceedings{AltunJH2011,
  title = {Loss Functions and Optimization Methods for Discriminative Learning of Label Sequences },
  author = {Altun, Y. and Johnson, M. and Hofmann, T.},
  pages = {145-152},
  editors = {Collins, M. , M. Steedman},
  publisher = {ACL},
  address = {East Stroudsburg, PA, USA},
  month = jul,
  year = {2003},
  doi = {},
  month_numeric = {7}
}