본문 바로가기

I'm Lim

검색하기
I'm Lim
프로필사진 imlim

  • 분류 전체보기 (56)
    • Deep Learning (19)
      • Fundamental (7)
      • Optimization (5)
      • Normalization (1)
      • Weight Initialization (6)
      • Terminology (0)
    • Classification (26)
      • Paper (3)
      • Implementation (1)
    • Object Detection (9)
    • Lecture & etc (2)
      • CS231n (0)
      • Git (1)
      • Torch (1)
Guestbook
Notice
Recent Posts
Link
Tags
  • SPP-Net
  • deep learning
  • LeNet 구현
  • Weight initialization
  • Convolution 종류
  • Optimizer
  • overfeat
  • image classification
  • object detection
  • 딥러닝
more
«   2026/04   »
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
Archives
Today
Total
관리 메뉴
  • 글쓰기
  • 방명록
  • RSS
  • 관리

목록DenseNet (1)

I'm Lim

[논문 정리] DenseNet

Paper Huang, Gao, et al. "Densely connected convolutional networks." Proceedings of the IEEE conference on computer vision and pattern recognition. 2017. Abstract ResNet 논문에서 레이어 간 shortcut connection을 이용하였을 때 더욱 안정적이고, 빠르게 학습 함을 보였다. ResNet이 레이어 간의 shotcut connection을 가지는 구조라면, DenseNet은 이전 모든 레이어와 현재 레이어에 shortcut connection을 가지는 구조다. 따라서, ResNet의 shortcut connection의 개수가 $L$개라면, DenseNet은 $\d..

Classification 2022. 11. 26. 12:59
Prev 1 Next

Blog is powered by AXZ / Designed by Tistory

티스토리툴바