Notice
Recent Posts
Recent Comments
Link
| 일 | 월 | 화 | 수 | 목 | 금 | 토 |
|---|---|---|---|---|---|---|
| 1 | ||||||
| 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| 9 | 10 | 11 | 12 | 13 | 14 | 15 |
| 16 | 17 | 18 | 19 | 20 | 21 | 22 |
| 23 | 24 | 25 | 26 | 27 | 28 | 29 |
| 30 |
Tags
- 의미연결망
- sequential data
- sensibleness
- Google Cloud Platform
- Holdout
- dynamic topic modeling
- ROC-AUC Curve
- 분류모델평가
- word representation
- degree centrality
- semantic network
- 감성분석
- 토픽모델링
- 동적토픽모델링
- type-hint
- 허깅페이스
- hugging face
- 구글클라우드플랫폼
- 알파베타가지치기
- GCP
- Min-Max 알고리즘
- 임베딩
- QANet
- sbert
- 머신러닝
- topic modeling
- word2vec
- 사회연결망분석
- Meena
- Enriching Word Vectors with Subword Information
Archives
- Today
- Total
목록fastText (1)
Dev.log
본 포스팅에서는 FastText(Enriching Word Vectors with Subword Information)논문에 대해 리뷰해 보도록 하겠습니다. FastText는 2017년 ACL에서 발표되었으며, 오늘날 워드임베딩의 개념인 word vectors representation을 학습시키는 방법에 대한 논문입니다. 본 논문이 등장 할 때 까지만 해도 워드 임베딩(word embedding)이라는 단어 대신 distributed 혹은 continious word represnetation이라는 단어가 주로 사용되었습니다. FastText은 Word2vec을 타겟 논문으로 잡았으며, word2vec의 몇가지 한계점에 대해 언급했습니다. Word2Vec의 한계점 Word2Vec의 한계는 하나의 단어에..
논문리뷰
2022. 3. 4. 10:41