Sciweavers

COLING
2008

Random Restarts in Minimum Error Rate Training for Statistical Machine Translation

14 years 27 days ago
Random Restarts in Minimum Error Rate Training for Statistical Machine Translation
Och's (2003) minimum error rate training (MERT) procedure is the most commonly used method for training feature weights in statistical machine translation (SMT) models. The use of multiple randomized starting points in MERT is a well-established practice, although there seems to be no published systematic study of its benefits. We compare several ways of performing random restarts with MERT. We find that all of our random restart methods outperform MERT without random restarts, and we develop some refinements of random restarts that are superior to the most common approach with regard to resulting model quality and training time.
Robert C. Moore, Chris Quirk
Added 29 Oct 2010
Updated 29 Oct 2010
Type Conference
Year 2008
Where COLING
Authors Robert C. Moore, Chris Quirk
Comments (0)