Riemannian Optimization for Skip-Gram Negative Sampling
Computation and Language
2017-04-27 v1
Abstract
Skip-Gram Negative Sampling (SGNS) word embedding model, well known by its implementation in "word2vec" software, is usually optimized by stochastic gradient descent. However, the optimization of SGNS objective can be viewed as a problem of searching for a good matrix with the low-rank constraint. The most standard way to solve this type of problems is to apply Riemannian optimization framework to optimize the SGNS objective over the manifold of required low-rank matrices. In this paper, we propose an algorithm that optimizes SGNS objective using Riemannian optimization and demonstrates its superiority over popular competitors, such as the original method to train SGNS and SVD over SPPMI matrix.
Cite
@article{arxiv.1704.08059,
title = {Riemannian Optimization for Skip-Gram Negative Sampling},
author = {Alexander Fonarev and Oleksii Hrinchuk and Gleb Gusev and Pavel Serdyukov and Ivan Oseledets},
journal= {arXiv preprint arXiv:1704.08059},
year = {2017}
}
Comments
9 pages, 4 figures, ACL 2017