A Locally Linear Procedure for Word Translation
Abstract
Learning a mapping between word embeddings of two languages given a dictionary is an important problem with several applications. A common mapping approach is using an orthogonal matrix. The Orthogonal Procrustes Analysis (PA) algorithm can be applied to find the optimal orthogonal matrix. This solution restricts the expressiveness of the translation model which may result in sub-optimal translations. We propose a natural extension of the PA algorithm that uses multiple orthogonal translation matrices to model the mapping and derive an algorithm to learn these multiple matrices. We achieve better performance in a bilingual word translation task and a cross lingual word similarity task compared to the single matrix baseline. We also show how multiple matrices can model multiple senses of a word.
Related Papers
- → The organisation and management of agricultural extension and farmer-assistance—A note on developments in Botswana. Part 1: Organisation for extension and some extension problems(1980)2 cited
- Study of n-dimensional Extension Set(2001)
- → An appraisal of the agriculture extension service in iran as perceived by extension specialists and extension agents(1982)1 cited
- On the Intension and Extension of “Brand Extension”(2003)