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 |
Tags
- flame
- Deeplearning
- Python
- 자료구조
- center loss
- cv2
- reconstruction
- transformer
- 스택
- Knowledge Distillation
- attention
- 큐
- Computer Vision
- 3D
- 알고리즘
- point cloud
- deep learning
- OpenCV
- numpy
- NLP
- 파이썬
- 임계처리
- 논문 구현
- Threshold
- re-identification
- Object Tracking
- level2
- 프로그래머스
- 딥러닝
- Object Detection
Archives
- Today
- Total
목록neural network approximation (1)
공돌이 공룡의 서재
![](http://i1.daumcdn.net/thumb/C150x150/?fname=https://blog.kakaocdn.net/dn/Fz35p/btre1au7Lj1/X3p2Msg8Sao4cecr1GCRy1/img.png)
XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks https://arxiv.org/pdf/1603.05279.pdf Quantization의 아버지 격 되는 논문이다. 인용수가 3천을 넘어간다. 아이디어 자체는 간단해서 수식적인 부분 외에는 수월하게 읽혔다. Introduction CNN 기반 모델들은 memory나 computational power가 많이 필요하다. 빠른 속도를 내려면 GPU 가 필요하기도 하고, 핸드폰이나 임베디드 전자 제품 같은 작은 기기들에는 적합하지 않다. 논문에서는 이에 대한 예시로 AlexNet과 VGG의 parameter 수와 메모리 크기를 설명하고 있다. 이 논문에서는 이에 대한 해결..
딥러닝
2021. 9. 14. 22:51