Today, neural network and deep learning are more and more popular but what materials to learn will create a little headache due to the huge numbers of resources online. Different people have different tastes. Someones like online tutorials, some take Coursera courses. For me, I prefer learning neural networks and its application from courses of some big universities (e.g. Stanford, Berkeley, CMU). I like courses with slides, lecture notes, lecture record and homeworks. For this reason, I summarize these courses, categorize them according to their difficulty; I try to keep the post updated.


In this section, I will list some courses that provide foundation for neural network learning.

  1. Designing, Visualizing and Understanding Deep Neural Networks from UC Berkeley.
  2. Stats 202: Data Mining and Analysis from Stanford. They also include some solutions for homework.


After learning foundation about neural network, we move to some of its applications

  1. CS224n: Natural Language Processing with Deep Learning from Stanford (Winter 2019). The note and materials are fantastic. Chris Manning is also a big name in the field. Must try course.
  2. CS231n: Convolutional Neural Networks for Visual Recognition from Stanford (2016). An other masterpiece. It will help you understand the whole neural network application for computer vision.


We move forward to more advanced technique of deep learning.

  1. CS 285 Deep Reinforcement Learning from Berkeley. It helps us to understand deep RL. There is another course from Stanford but I feel that the materials of Stanford are not that details to follow.
  2. Statistical Learning from Stanford. It is so advanced, be careful.
  3. Convex Optimization. The instructor is Stephen Boyd. He is a god in this field. Follow the sequence if you enjoy the course.


Some useful materials for neural network

  1. The Annotated Transformer from Harvard. It is for us to understand Transformer - an attention neural network.
  2. How to Escape Saddle Points Efficiently from Berkeley. I think the title says all, I dont need to explain :)

Hopefully, these courses are useful for your study.