Loss in Translation: Learning Bilingual Word Mapping with a Retrieval Criterion

Armand Joulin, Piotr Bojanowski, Tomas Mikolov, Hervé Jégou, Edouard Grave


Abstract
Continuous word representations learned separately on distinct languages can be aligned so that their words become comparable in a common space. Existing works typically solve a quadratic problem to learn a orthogonal matrix aligning a bilingual lexicon, and use a retrieval criterion for inference. In this paper, we propose an unified formulation that directly optimizes a retrieval criterion in an end-to-end fashion. Our experiments on standard benchmarks show that our approach outperforms the state of the art on word translation, with the biggest improvements observed for distant language pairs such as English-Chinese.
Anthology ID:
D18-1330
Volume:
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
Month:
October-November
Year:
2018
Address:
Brussels, Belgium
Venue:
EMNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
2979–2984
Language:
URL:
https://aclanthology.org/D18-1330
DOI:
10.18653/v1/D18-1330
Bibkey:
Cite (ACL):
Armand Joulin, Piotr Bojanowski, Tomas Mikolov, Hervé Jégou, and Edouard Grave. 2018. Loss in Translation: Learning Bilingual Word Mapping with a Retrieval Criterion. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2979–2984, Brussels, Belgium. Association for Computational Linguistics.
Cite (Informal):
Loss in Translation: Learning Bilingual Word Mapping with a Retrieval Criterion (Joulin et al., EMNLP 2018)
Copy Citation:
PDF:
https://aclanthology.org/D18-1330.pdf
Attachment:
 D18-1330.Attachment.zip
Code
 additional community code