循环神经网络理论到实践(1)

1. 导读

循环神经网络是在NLP（自然语言处理）领域最常使用的神经网络结构，和卷积神经网络在图像识别领域的地位类似。之前我们用4篇博文，详细介绍了CNN，现在，在使用CNN进行各种实践之前，有必要再介绍RNN.之后再综合实践，会更好一些。
一些有价值的链接：
http://www.cnblogs.com/neopenx/p/4623328.html
https://yjango.gitbooks.io/superorganism/content/%E9%80%92%E5%BD%92%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C.html
https://wugh.github.io/posts/2016/03/cs224d-notes4-recurrent-neural-networks/?utm_source=tuicool&utm_medium=referral
https://wugh.github.io/posts/2016/05/cs224d-notes5-recusive-neural-networks/
https://www.zhihu.com/question/36824148
https://zhuanlan.zhihu.com/p/22266022
http://www.wildml.com/2015/09/recurrent-neural-networks-tutorial-part-1-introduction-to-rnns/
https://www.tensorflow.org/tutorials/recurrent
http://blog.youkuaiyun.com/heyongluoyao8/article/details/48636251
https://clyyuanzi.gitbooks.io/julymlnotes/content/dl_rnn.html
https://zybuluo.com/hanbingtao/note/541458
http://www.cnblogs.com/neopenx/p/4623328.html