2018
ACL
ACL 2018
Multi-representation ensembles and delayed SGD updates improve syntax-based NMT
Abstract
AbstractWe explore strategies for incorporating target syntax into Neural Machine Translation. We specifically focus on syntax in ensembles containing multiple sentence representations. We formulate beam search over such ensembles using WFSTs, and describe a delayed SGD update training procedure that is especially effective for long representations like linearized syntax. Our approach gives state-of-the-art performance on a difficult Japanese-English task.
🌉
Interdisciplinary Bridge
- Artificial Intelligence and Natural Language Processing
Topics
Machine Learning > Core Methods > Representation Learning
Natural Language Processing > Understanding > Syntax
Natural Language Processing > Applications > Machine Translation
Machine Learning > Core Methods > Ensemble Methods
Deep Learning > Learning Types > Deep Learning
Artificial Intelligence > Core AI > Language
Deep Learning > Learning Types > Ensemble Learning