Characterizing Departures from Linearity in Word Translation

Ndapa Nakashole, Raphael Flauger


Abstract
We investigate the behavior of maps learned by machine translation methods. The maps translate words by projecting between word embedding spaces of different languages. We locally approximate these maps using linear maps, and find that they vary across the word embedding space. This demonstrates that the underlying maps are non-linear. Importantly, we show that the locally linear maps vary by an amount that is tightly correlated with the distance between the neighborhoods on which they are trained. Our results can be used to test non-linear methods, and to drive the design of more accurate maps for word translation.
Anthology ID:
P18-2036
Volume:
Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Month:
July
Year:
2018
Address:
Melbourne, Australia
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
221–227
Language:
URL:
https://aclanthology.org/P18-2036
DOI:
10.18653/v1/P18-2036
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/P18-2036.pdf