arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.07013cs.CLcs.AI
收藏

解构词嵌入算法

Deconstructing word embedding algorithms

BMO AI Capabilities Team(BMO人工智能能力团队);Bank of Montreal(蒙特利尔银行);Mila - Québec AI Institute(Mila-魁北克人工智能研究所);McGill University(麦吉尔大学)

Kian Kenyon-Dean, Edward Newell, Jackie Chi Kit Cheung

摘要:

词嵌入是词的可靠特征表示,用于获得各种NLP应用的高质量结果。无语境的词嵌入在当今的许多NLP任务中使用,特别是在高内存容量和GPU不可用的资源有限的设置中。鉴于词嵌入在NLP中的历史成功,我们提出了一些最知名的词嵌入算法的回顾。在这项工作中,我们解构Word2vec,GloVe,和其他人,到一个共同的形式,揭示了一些共同的条件,似乎是所需的性能词嵌入。我们相信,本文的理论研究结果可以为未来模型的更明智的发展提供基础。

英文摘要:

Word embeddings are reliable feature representations of words used to obtain high quality results for various NLP applications. Uncontextualized word embeddings are used in many NLP tasks today, especially in resource-limited settings where high memory capacity and GPUs are not available. Given the historical success of word embeddings in NLP, we propose a retrospective on some of the most well-known word embedding algorithms. In this work, we deconstruct Word2vec, GloVe, and others, into a common form, unveiling some of the common conditions that seem to be required for making performant word embeddings. We believe that the theoretical findings in this paper can provide a basis for more informed development of future models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑