300x250
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 |
Tags
- 리눅스
- 딥러닝
- BFS
- 백준
- 동적 프로그래밍
- nlp
- 파이썬
- PYTHON
- DFS
- 코딩테스트
- kick start
- 프로그래밍
- 구글 킥스타트
- 운영체제
- 알고리즘
- dp
- 순열
- 킥스타트
- CSS
- 그래프
- 프로그래머스
- OS
- AI
- linux
- 코딩
- 브루트포스
- 코딩 테스트
- 동적프로그래밍
- google coding competition
- 네트워크
Archives
- Today
- Total
목록fine tunining (1)
오뚝이개발자
BERT(NAACL-HLT 2019) 논문 리뷰
BERT 원논문 링크 What is BERT? Bidirectional Encoder Representations from Transformers Can be fine-tuned with jus one additional output layer for a wide range of task Don't need task-specific architecture Motive BERT는 left to right LM이 sentence level task에서 sub-optimal이라는 점을 지적 기존 LTR LM은 주로 양방향으로부터의 문맥정보를 incorporate하는 것이 중요한 경우 취약(e.g. question answering) Transformer에서 착안한 BERT Encoder Attention을 사..
AI/Deep Learning Paper Review
2021. 10. 8. 14:25