The University of Edinburgh’s Neural MT Systems for WMT17
| Autor: | Antonio Valerio Miceli Barone, Alexandra Birch, Barry Haddow, Ulrich Germann, Kenneth Heafield, Philip Williams, Rico Sennrich, Anna Currey |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Předmět: |
FOS: Computer and information sciences
Normalization (statistics) Czech Machine translation Turkish Computer science 02 engineering and technology computer.software_genre Task (project management) German 030507 speech-language pathology & audiology 03 medical and health sciences 0202 electrical engineering electronic engineering information engineering Computer Science - Computation and Language business.industry Romanian Latvian language.human_language language 020201 artificial intelligence & image processing Artificial intelligence 0305 other medical science business Computation and Language (cs.CL) computer Natural language processing |
| Zdroj: | Sennrich, R, Birch, A, Currey, A, Germann, U, Haddow, B, Heafield, K, Miceli Barone, A V & Williams, P 2017, The University of Edinburgh’s Neural MT Systems for WMT17 . in Proceedings of the Second Conference on Machine Translation . pp. 389-399, Second Conference on Machine Translation, Copenhagen, Denmark, 7/09/17 . https://doi.org/10.18653/v1/W17-4739 WMT Proceedings of the Second Conference on Machine Translation |
| Popis: | This paper describes the University of Edinburgh's submissions to the WMT17 shared news translation and biomedical translation tasks. We participated in 12 translation directions for news, translating between English and Czech, German, Latvian, Russian, Turkish and Chinese. For the biomedical task we submitted systems for English to Czech, German, Polish and Romanian. Our systems are neural machine translation systems trained with Nematus, an attentional encoder-decoder. We follow our setup from last year and build BPE-based models with parallel and back-translated monolingual training data. Novelties this year include the use of deep architectures, layer normalization, and more compact models due to weight tying and improvements in BPE segmentations. We perform extensive ablative experiments, reporting on the effectivenes of layer normalization, deep architectures, and different ensembling techniques. Comment: WMT 2017 shared task track; for Bibtex, see http://homepages.inf.ed.ac.uk/rsennric/bib.html#uedin-nmt:2017 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|