Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
文献紹介 7月16日
Search
gumigumi7
July 16, 2018
0
260
文献紹介 7月16日
Probabilistic FastText for Multi-Sense Word Embeddings
gumigumi7
July 16, 2018
Tweet
Share
More Decks by gumigumi7
See All by gumigumi7
文献紹介 1月24日
gumigumi7
0
230
文献紹介 11月7日
gumigumi7
0
120
文献紹介 10月3日
gumigumi7
0
320
文献紹介 9月3日
gumigumi7
0
250
文献紹介 8月10日
gumigumi7
0
120
文献紹介 6月12日
gumigumi7
0
330
文献紹介 5月16日
gumigumi7
0
180
文献紹介 4月18日
gumigumi7
0
140
文献紹介 12月15日
gumigumi7
0
110
Featured
See All Featured
Distributed Sagas: A Protocol for Coordinating Microservices
caitiem20
323
20k
The Power of CSS Pseudo Elements
geoffreycrofte
61
5k
Build The Right Thing And Hit Your Dates
maggiecrowley
25
2k
Building Adaptive Systems
keathley
32
1.9k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
10
1k
The Psychology of Web Performance [Beyond Tellerrand 2023]
tammyeverts
10
1.5k
Creatively Recalculating Your Daily Design Routine
revolveconf
211
11k
Done Done
chrislema
178
15k
Become a Pro
speakerdeck
PRO
12
4.6k
Atom: Resistance is Futile
akmur
260
25k
Mobile First: as difficult as doing things right
swwweet
217
8.6k
Agile that works and the tools we love
rasmusluckow
325
20k
Transcript
) ( Probabilistic FastText for Multi-Sense Word Embeddings
▪ ▪ Mikhail Khodak, Nikunj Saunshi, Yingyu Liang,
Tengyu Ma, Brandon Stewart, Sanjeev Arora. ▪ Probabilistic FastText for Multi-Sense Word Embeddings. ▪ Proceedings of the Association of Computational Linguistics. 2018. ▪ ▪ GaussianMixtureSense Embedding 2
▪ n o V e dT WG oe2 W
W ▪ V e a i M V a c ▪ V a n o 3
▪ e c V ▪ ▪ c V c
W ▪ e c ” ” ▪ fastText ▪ Word2Vec subwordc V c 2 ▪ Word2Vec e c ” ” 4
▪ emt a ▪ ( ) ▪ G ul
m emt er ksu a a u l m a emt ▪ emt M a S ▪ MUSE (Modularizing Unsupervised Sense Embeddings) ▪ api t api t a ▪ Contextual Word Similarity k e SoTA 5
▪ fa E u wn E g ue bd
▪ , B A A C A ▪ a g e bd G ▪ h mo ] V x ptk ▪ A , A C A C , ▪ ] V bdxMpt i pkLR D r ov cT W[ps 6
( ) ( ▪ 7
( ) ( ▪ 8 !
: "#,% : & ' (# : 1
( ) ( ▪ , ▪ 9
( ) ( ▪ ▪
10 Negative Sampling Margin
▪ 11
Subword
12 ▪
▪ RockBank
▪ SCWS-68 ▪ “... east bank of the Des
Moines River ...” ++ “... basis of all money laundering ...” +. bank + money . :*$'5. ()8 ▪ - bank + money 79)8. ",8 79)8'4.*/#97. //& ▪ -3'+!.1: -6()
15 ▪
16 ▪ $ Subwords+ ( %%' 0-.1 &)(%
▪ Abnormal Abnormality ! Subwords %"#+ ( *(
▪ a i rtx n e M sx ▪
e T T G ▪ F ea e e 17