本文是对ImageNet 2015的冠军ResNet(Deep Residual Networks)以及目前围绕ResNet这个工作研究者后续所发论文的总结,主要涉及到下面5篇论文。
1. Link: Highway Networks:
Rupesh Kumar Srivastava, Klaus Greff, Jürgen Schmidhuber
2. Link: Deep Residual Learning for Image Recognition:
Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun
3. Link: Identity Mappings in Deep Residual Networks:
Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun
4. Link: ResNet in ResNet: Generalizing Residual Architectures:
Sasha Targ, Diogo Almeida, Kevin Lyman
5. Link: Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning:
Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke
Highway Networks
Deep Residual Learning for Image Recognition
Identity Mappings in Deep Residual Networks
ResNet in ResNet: Generalizing Residual Architectures
Inception-v4, Inception-ResNet
因为出了ResNet,所以Google的researcher马上改造了Inception-v3的网络,引入了类似ResNet block的结构,实验证明,效果也确实变好了,文章就不过多解读了。
本文标题:Re-thinking Deep Residual Networks
文章作者:Binbin Xu
发布时间:2016年03月30日 - 19时21分
最后更新:2016年03月31日 - 15时18分
原始链接:http://freesouls.github.io/2016/03/30/rethinking-deep-residual-networks/
许可协议: “署名-非商用-相同方式共享 3.0” 转载请保留原文链接及作者。