Work

Dynamic Data Selection for Neural Machine Translation

Year: 2017

Type: conference-paper

Abstract: 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.

Source: Conference on Empirical Methods in Natural Language Processing (EMNLP)

DOI: 10.18653/v1/d17-1147

Authors: Marlies van der Wees, Arianna Bisazza, Christof Monz

Institutions: University of Amsterdam, Leiden University

Language: English

Topic: Natural Language Processing Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funder: Nederlandse Organisatie voor Wetenschappelijk Onderzoek

Award: Connecting the Dots: Minimal Structure Modeling for Machine Translation