Download | - View accepted manuscript: Batch tuning strategies for statistical machine translation (PDF, 595 KiB)
|
---|
Author | Search for: Cherry, Colin1; Search for: Foster, George1 |
---|
Affiliation | - National Research Council of Canada. Information and Communication Technologies
|
---|
Format | Text, Article |
---|
Conference | North American Chapter of the Association for Computational Linguistics: Human Language Technologies - (HLT-NAACL 2012), Montreal, QC, June 3-8, 2012 |
---|
Abstract | There has been a proliferation of recent work on SMT tuning algorithms capable of handling larger feature sets than the traditional MERT approach. We analyze a number of these algorithms in terms of their sentence-level loss functions, which motivates several new approaches, including a Structured SVM. We perform empirical comparisons of eight different tuning strategies, including MERT, in a variety of settings. Among other results, we find that a simple and efficient batch version of MIRA performs at least as well as training online, and consistently outperforms other options. |
---|
Publication date | 2012-06-08 |
---|
Publisher | Association for Computational Linguistics |
---|
In | |
---|
Language | English |
---|
Peer reviewed | Yes |
---|
NPARC number | 20262877 |
---|
Export citation | Export as RIS |
---|
Report a correction | Report a correction (opens in a new tab) |
---|
Record identifier | 1101df04-9f92-4758-a257-3a8457183e06 |
---|
Record created | 2012-07-10 |
---|
Record modified | 2020-04-21 |
---|