Dynamic Data Selection for Neural Machine Translation
| Autor: | Arianna Bisazza, Christof Monz, Marlies van der Wees |
|---|---|
| Přispěvatelé: | IvI Research (FNWI), Information and Language Processing Syst (IVI, FNWI) |
| Rok vydání: | 2017 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Computation and Language Phrase Training set Machine translation business.industry Computer science Dynamic data 02 engineering and technology computer.software_genre Translation (geometry) Machine learning 030507 speech-language pathology & audiology 03 medical and health sciences 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence 0305 other medical science business Computation and Language (cs.CL) computer Selection (genetic algorithm) BLEU |
| Zdroj: | EMNLP Conference on Empirical Methods in Natural Language Processing: emnlp20017 : Copenhagen, Denmark, September 7-11, 2017 : conference proceedings, 1400-1410 STARTPAGE=1400;ENDPAGE=1410;TITLE=Conference on Empirical Methods in Natural Language Processing |
| DOI: | 10.18653/v1/d17-1147 |
| Popis: | Intelligent selection of training data has proven a successful technique to simultaneously increase training efficiency and translation performance for phrase-based machine translation (PBMT). With the recent increase in popularity of neural machine translation (NMT), we explore in this paper to what extent and how NMT can also benefit from data selection. While state-of-the-art data selection (Axelrod et al., 2011) consistently performs well for PBMT, we show that gains are substantially lower for NMT. Next, we introduce dynamic data selection for NMT, a method in which we vary the selected subset of training data between different training epochs. Our experiments show that the best results are achieved when applying a technique we call gradual fine-tuning, with improvements up to +2.6 BLEU over the original data selection approach and up to +3.1 BLEU over a general baseline. Accepted at EMNLP2017 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|