Wei Zhu's DL Station | 又一个入坑的DL researcher
Coursera: Neural Networks for Machine Learning- Lecture 7

View more

Coursera: Neural Networks for Machine Learning- Lecture 7

Modeling sequences & RNN

Coursera: Neural Networks for Machine Learning- Lecture 6

View more

Coursera: Neural Networks for Machine Learning- Lecture 6

Overview of mini-batch gradient descent & tricks

Coursera: Neural Networks for Machine Learning- Lecture 5

View more

Coursera: Neural Networks for Machine Learning- Lecture 5

Object recognition & CNN

Coursera: Neural Networks for Machine Learning- Lecture 4

View more

Coursera: Neural Networks for Machine Learning- Lecture 4

Language Models & Softmax output

Coursera: Neural Networks for Machine Learning- Lecture 3

View more

Coursera: Neural Networks for Machine Learning- Lecture 3

Learning procedure of Neural Networks

Coursera: Neural Networks for Machine Learning- Lecture 2

View more

Coursera: Neural Networks for Machine Learning- Lecture 2

An overview of the main types of neural networks architecture

Coursera: Neural Networks for Machine Learning- Lecture 1

View more

Coursera: Neural Networks for Machine Learning- Lecture 1

学过神经网络并且用过Coursera的人应该都看过这个课程,它是由Hinton主讲,深度学习的开山鼻祖,他的英文听着有点别扭,而且他的课程包括文章都有点晦涩难懂,在这方面Andrew Ng就好太多了,但是牛人可能就是这样,就算再简单的东西也得说的复杂点不是。。。我第一次学这门课也处于半懂不懂的状态,反正就这么看完了。但是看了几个月文献以后,我又想起来再回头复习一下,虽然得花不少时间,但这时候再回头看确实受益匪浅,以下就是我贴出来的主要一些ppt页,另外附加了点注释,希望对大家有帮助。

1998_Efficient Backprop

View more

1998_Efficient Backprop

看了神经网络许久,开始自己训练一些网络,用的只是deeplearning toolbox中的一些样例,发现在实际过程中,训练网络达到最优解需要很多的tricks,实际上神经网络的调参比理论更难,记得某篇文章中说过,调参实际上more than the art of theory,这篇文章是1998年Yann LeCun的应用文章,讲解了许多调参经验,对初学者帮助很大。

Deep big simple neural nets excel on hand-written digit recognition

View more

Deep big simple neural nets excel on hand-written digit recognition

2010_Deep big simple neural nets excel on hand-written digit recognition 首先看这篇文章: 2003_Best practice for convolutional neural networks applied to visual document analysis ...
An analysis of single-layer network in unsupervised feature learning

View more

An analysis of single-layer network in unsupervised feature learning

此文详细研究了单层网络的几个因素,分析很到位,对初学者帮助较大。