深度学习资料汇总

1. 入门

    Neural Network(别名)：Deep Learning

    Deep Learning比较流行的一个原因，主要是因为它能够自主地从数据上学到有用的feature。特别是对于一些不知道如何设计feature的场合，比如说图像和speech。deep learning可以学习到比以往比如说sift或者MFCC这样手工设计的feature更好的方法，而且这些feature有很强的semantic的含义。所以很多时候在用到其他的一些task的时候会很有效，这也是为什么我们可以用一个feature来实现很多比如说识别，检测，物体分割这样的不同task的缘故。

    1)  Machine Learning Mastery(初学者、从业人员、高级人员)      

    2)  CS231n: Convolutional Neural Networks for Visual Recognition (Standford: learn to implement, train and debug)

    3)  Python Numpy Tutorial

    4)  Neural Networks and Deep Learning

    5)  Neural Networks for Machine Learning (Geoffrey Hinton)

    6)  Machine Learning: 2014-2015(Nando de Freitas Oxford)

    7)  CS224d: Deep Learning for Natural Language Processing 

    8) Colah博客，非常浅显精彩的DL入门

    9) Deep Learning Summer School, Montreal 2015(Deep Learning，Unsupervised Learning，Reinforcement Learning)

    10) Yann LeCun 2015-2016 Deep Learning

    11) Deep Learning for Computer Vision Barcelona

    12) Computer Vision (CS 543 / ECE 549) – Spring 2015

    13) CS3710: Visual Recognition, Spring 2015

    13) 16-385 Computer Vision 2015

       

2. 进阶

      1) Deep Learning Book

    2) Recurrent Neural Networks

        - The Unreasonable Effectiveness of Recurrent Neural Networks
        - Neural Networks (General)
        - Understanding LSTM Networks

    3) Deep learning frameworks
        - Tensorflow (by Google) tutorials
        - Torch (by Facebook)
        - Theano (by MILA)

   4) Evaluation of Deep Learning Frameworks

   5)

3. Models

    1) Caffe Model Zoo: Lots of pretrained ConvNets 

    2) ConvNetJS CIFAR-10 demo

    3) Deep Residual Networks (ResNets)

4. Computer Vision

      1)  Computer Vision (CS 543 / ECE 549) – Spring 2015

      2) Awesome Computer Vision
      3) CSE/EE486 Computer Vision I (Stereo Matching)

      4)  CSE455: Computer Vision (Stereo Matching)

      5)  CS 664 Computer Vision – Spring 2008 (Stereo Matching)

      6)  CSCE 643 - Computer Vision(Jinxiang Chai，有很多资源链接) 

      7)  CSCE 441: Computer Graphics (Jinxiang Chai:vSpring 2016)

5. 强化学习

      1) CS 294: Deep Reinforcement Learning, Fall 2015          

6. Caffe实战

      1) Caffe + Ubuntu 15.04 + CUDA 7.5 新手安装配置指南

      2) Caffe上手教程

      3) Yangqing Jia  (Caffe作者，现由BVLC维护）

7. RNN

• recurrent neural nets augmented with differentiable memory modules (e.g. Memory Networks:
  • Memory Networks (FAIR): https://scholar.google.com/citat...
  • Stack-Augmented RNN (FAIR): Google Scholar Citations
  • Neural Turing Machine (DeepMind): [1410.5401] Neural Turing Machines
  • End-toEnd MemNN (FAIR/NYU): https://scholar.google.com/citat...

8. Hadoop和Spark DL工具

    1) Deeplearning4J  

9. 他人博客

    1) zouxy09-机器学习（基于概念）

10. Udacity开源数据

    1) 行驶数据集

    

    2) 自然场景短视频数据

    

11. SLAM        

     1) Google开源SLAM

     

12.  图片检测

    1) Yahoo开源色情图片检测

13. 对抗机器学习 (Adversarial Machine Learning)

    1) Ian Goodfellow开源对抗机器学习库cleverhans v0.1

14. Human Pose Estimation

    1) 2D : MPII Human Pose Models

参考：

1） http://ofir.io/How-to-Start-Learning-Deep-Learning/

2）