Representations of Words and Phrases and their Compositionality: Difference between revisions
Jump to navigation
Jump to search
Line 10: | Line 10: | ||
The Skip-gram model is a method of constructing a Word2Vec encoding using a neural network. Individual words are encoded | The Skip-gram model is a method of constructing a Word2Vec encoding using a neural network. Individual words are encoded | ||
[[File: | [[File:skipmod.JPG|200px|thumb|left|alt text]] | ||
= Skip Gram Model = | = Skip Gram Model = |
Revision as of 01:24, 21 November 2018
Representations of Words and Phrases and their Compositionality is a popular paper published by the Google team led by Tomas Mikolov in 2013. It is known for its impact in the field of Natural Language Processing and the techniques described below are till in practice today.
Presented by
- F. Jiang
- J. Hu
- Y. Zhang
Introduction
The Skip-gram model is a method of constructing a Word2Vec encoding using a neural network. Individual words are encoded
Skip Gram Model
Hierarchical Softmax
Negative Sampling
Subsampling of Frequent Words
Empirical Results
References
[1] Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg Corrado, Jeffrey Dean (2013). Distributed Representations of Words and Phrases and their Compositionality. arXiv:1310.4546.
[2] McCormick, C. (2017, January 11). Word2Vec Tutorial Part 2 - Negative Sampling. Retrieved from http://www.mccormickml.com