Unsupervised Hyper-alignment for Multilingual Word Embeddings

Anonymous

Sep 27, 2018 (modified: Oct 10, 2018) ICLR 2019 Conference Blind Submission readers: everyone Show Bibtex
  • Abstract: We consider the problem of aligning continuous word representations, learned in multiple languages, to a common space. It was recently shown that, in the case of two languages, it is possible to learn such a mapping without supervision. In this paper, we extend one of the proposed methods to the problem of aligning multiple languages to a common space. A simple solution to this problem is to independently map all languages to English. Unfortunately, this can degrade the alignments between languages different from English. We thus propose to add constraints to ensure that the learned mappings can be composed, leading to better alignments. We evaluate our method on the problem of aligning word vectors in eleven languages, showing improvement in word translation requiring the composition of mappings.
0 Replies

Loading