This paper explores an incremental training strategy for the skip-gram model with negative sampling (SGNS) from both empirical and theoretical perspectives. Existing methods of neural word embeddings, including SGNS, are multi-pass algorithms and thus cannot perform incremental model update. To address this problem, we present a simple incremental extension of SGNS and provide a thorough theoretical analysis to demonstrate its validity. Empirical experiments demonstrated the correctness of the theoretical analysis as well as the practical usefulness of the incremental algorithm.
@article{arxiv.1704.03956,
title = {Incremental Skip-gram Model with Negative Sampling},
author = {Nobuhiro Kaji and Hayato Kobayashi},
journal= {arXiv preprint arXiv:1704.03956},
year = {2017}
}