https://github.com/cmudeeplearning11785/deep-learning-tutorials/blob/master/recitation-3/recitation3.ipynb
https://blog.csdn.net/shuzfan/article/details/51338178
https://zhuanlan.zhihu.com/p/22028079
https://github.com/cmudeeplearning11785/deep-learning-tutorials/blob/master/recitation-3/recitation3.ipynb
solution:
https://blog.csdn.net/tsq292978891/article/details/78619384
adam algorithm from paper
5min to finish
↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓↓
https://zhuanlan.zhihu.com/p/23178423
受不了这个习题课小哥的紧张了（后期完全口吃，听不懂他的英语。。。。）
https://www.zhihu.com/question/38102762
Must Know Tips/Tricks in Deep Neural Networks (by Xiu-Shen Wei)
http://lamda.nju.edu.cn/weixs/project/CNNTricks/CNNTricks.html
贝叶斯优化：
http://www.360doc.com/content/17/0810/05/43535834_678049865.shtml
https://www.youtube.com/watch?v=xP5dBw5cewg：
不带噪声和带噪声的公式
