Classical papers on CNN

less than 1 minute read

Published:

  • AlexNet (2012)
    1. 5个卷积块+3个全连接,建立了CNN的标准流程。
    2. 加速策略:模型并行。
  • ZF Net (2013)
    1. 可视化和理解神经网络。CNN领域可视化理解的开山之作,这篇文献告诉我们CNN的每一层到底学习到了什么特征,然后作者通过可视化进行调整网络,提高了精度。
    2. DeConvNet
  • VGG (2014)
    1. 相当经典!

Reference:

  1. https://adeshpande3.github.io/The-9-Deep-Learning-Papers-You-Need-To-Know-About.html
  2. https://blog.csdn.net/fu6543210/article/details/80407911
  3. https://www.tinymind.net.cn/articles/3dedb66b996232
  4. https://hackmd.io/@machine-learning/ByaTE80BI#ZFNetDeconvNet-Summary-and-Implementation

Tags:

Viewed times

Leave a Comment