Scalable Purely-Discriminative Training for Word and Tree Transducers

Benjamin Wellington, Joseph Turian, Chris Pike, Dan Melamed


Abstract
Discriminative training methods have recently led to significant advances in the state of the art of machine translation (MT). Another promising trend is the incorporation of syntactic information into MT systems. Combining these trends is difficult for reasons of system complexity and computational complexity. The present study makes progress towards a syntax-aware MT system whose every component is trained discriminatively. Our main innovation is an approach to discriminative learning that is computationally efficient enough for large statistical MT systems, yet whose accuracy on translation sub-tasks is near the state of the art. Our source code is downloadable from http://nlp.cs.nyu.edu/GenPar/.
Anthology ID:
2006.amta-papers.28
Volume:
Proceedings of the 7th Conference of the Association for Machine Translation in the Americas: Technical Papers
Month:
August 8-12
Year:
2006
Address:
Cambridge, Massachusetts, USA
Venue:
AMTA
SIG:
Publisher:
Association for Machine Translation in the Americas
Note:
Pages:
251–260
Language:
URL:
https://aclanthology.org/2006.amta-papers.28
DOI:
Bibkey:
Cite (ACL):
Benjamin Wellington, Joseph Turian, Chris Pike, and Dan Melamed. 2006. Scalable Purely-Discriminative Training for Word and Tree Transducers. In Proceedings of the 7th Conference of the Association for Machine Translation in the Americas: Technical Papers, pages 251–260, Cambridge, Massachusetts, USA. Association for Machine Translation in the Americas.
Cite (Informal):
Scalable Purely-Discriminative Training for Word and Tree Transducers (Wellington et al., AMTA 2006)
Copy Citation:
PDF:
https://aclanthology.org/2006.amta-papers.28.pdf