@inproceedings{mimno-thompson-2017-strange,
title = "The strange geometry of skip-gram with negative sampling",
author = "Mimno, David and
Thompson, Laure",
editor = "Palmer, Martha and
Hwa, Rebecca and
Riedel, Sebastian",
booktitle = "Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing",
month = sep,
year = "2017",
address = "Copenhagen, Denmark",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/D17-1308",
doi = "10.18653/v1/D17-1308",
pages = "2873--2878",
abstract = "Despite their ubiquity, word embeddings trained with skip-gram negative sampling (SGNS) remain poorly understood. We find that vector positions are not simply determined by semantic similarity, but rather occupy a narrow cone, diametrically opposed to the context vectors. We show that this geometric concentration depends on the ratio of positive to negative examples, and that it is neither theoretically nor empirically inherent in related embedding algorithms.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mimno-thompson-2017-strange">
<titleInfo>
<title>The strange geometry of skip-gram with negative sampling</title>
</titleInfo>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Mimno</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Laure</namePart>
<namePart type="family">Thompson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2017-09</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Martha</namePart>
<namePart type="family">Palmer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rebecca</namePart>
<namePart type="family">Hwa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sebastian</namePart>
<namePart type="family">Riedel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Copenhagen, Denmark</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Despite their ubiquity, word embeddings trained with skip-gram negative sampling (SGNS) remain poorly understood. We find that vector positions are not simply determined by semantic similarity, but rather occupy a narrow cone, diametrically opposed to the context vectors. We show that this geometric concentration depends on the ratio of positive to negative examples, and that it is neither theoretically nor empirically inherent in related embedding algorithms.</abstract>
<identifier type="citekey">mimno-thompson-2017-strange</identifier>
<identifier type="doi">10.18653/v1/D17-1308</identifier>
<location>
<url>https://aclanthology.org/D17-1308</url>
</location>
<part>
<date>2017-09</date>
<extent unit="page">
<start>2873</start>
<end>2878</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The strange geometry of skip-gram with negative sampling
%A Mimno, David
%A Thompson, Laure
%Y Palmer, Martha
%Y Hwa, Rebecca
%Y Riedel, Sebastian
%S Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing
%D 2017
%8 September
%I Association for Computational Linguistics
%C Copenhagen, Denmark
%F mimno-thompson-2017-strange
%X Despite their ubiquity, word embeddings trained with skip-gram negative sampling (SGNS) remain poorly understood. We find that vector positions are not simply determined by semantic similarity, but rather occupy a narrow cone, diametrically opposed to the context vectors. We show that this geometric concentration depends on the ratio of positive to negative examples, and that it is neither theoretically nor empirically inherent in related embedding algorithms.
%R 10.18653/v1/D17-1308
%U https://aclanthology.org/D17-1308
%U https://doi.org/10.18653/v1/D17-1308
%P 2873-2878
Markdown (Informal)
[The strange geometry of skip-gram with negative sampling](https://aclanthology.org/D17-1308) (Mimno & Thompson, EMNLP 2017)
ACL