Difference between revisions of "Hopkins and May, EMNLP 2011. Tuning as Ranking"
From Cohen Courses
Jump to navigationJump to searchLine 6: | Line 6: | ||
== Summary == | == Summary == | ||
− | This [[Category::paper]] | + | This [[Category::paper]] presents a simple and scalable method for statistical machine translation parameter tuning based on the pairwise approach to ranking. This algorithm has advantages over [[UsesMethod:: minimum error rate training | MERT]] [[RelatedPaper::Och, 2003]] |
== Method == | == Method == |
Revision as of 17:39, 29 November 2011
Contents
Citation
Mark Hopkins and Jonathan May. 2011. Tuning as Ranking. In Proceedings of EMNLP-2011.
Online Version
Summary
This paper presents a simple and scalable method for statistical machine translation parameter tuning based on the pairwise approach to ranking. This algorithm has advantages over MERT Och, 2003