Riemannian Optimization for Skip-Gram Negative Sampling
This work addresses the optimization challenge in word embeddings for natural language processing, offering an incremental improvement over standard techniques.
The paper tackles the optimization of the Skip-Gram Negative Sampling (SGNS) word embedding model by framing it as a low-rank matrix search problem and applying Riemannian optimization, resulting in a proposed algorithm that outperforms existing methods like the original SGNS training and SVD over SPPMI matrix.
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.