Reading directions and slides from videos

Relevant parts of Ian Goodfellow, Yoshua Bengio and Aaron Courville, Deep Learning, MIT Press, 2016, http://www.deeplearningbook.org (Links to an external site.),

  • Lecture 1 
    • Chapter 1: an introduction to deep learning and some history
    • Chapter 5-5.1, 5.7.1: machine learning basics.
    • Chapter 5.10: building a machine learning algorithm.
  • Lecture 2
    • Maximum likelihood estimation, Chapter 5.5.
    • Chapter 6-6.4: Deep feedforward networks.
    • Empirical risk minimization and surrogate losses, Chapter 8-8.1.2.
    • To understand the term cross-entropy (and learn about other important concepts): Chapter 3.13.
    • Gradient descent optimization: Chapter 4.3 (you can skip 4.3.1).
  • Lecture 3
    • Chapter 5.9, 6.5, 8.1.3, 8.3, 8.5
  • Lecture 4
    • Chapter 5.2, 5.3, 7.1 and 8.7
  • Lectures 5 and 6
    • Chapter 9, except 9.6 and 9.8-9.10

We don't have slides from the videos by Andrew Ng, but slides from most of the other videos are listed here: 

We believe it can be valuable to look at different explanations of the same material. We have therefore collected a list of resources that you may find interesting. Note that we do not expect you to look at them and that they are merely included for those who are interested.

Entire courses

YouTube

Deep learning is a popular topic and many people are producing instruction videos. Here is a selected list:

Transformers

Even though transformers is beyond the scope of this module, it is reasonable to mention them as an important alternative (or complement) to CNNs for computer vision. Below are a small number of references that you may find inspiring. 

Even though many recent papers on computer vision involve transformers, most of them also make use of CNNs. Learning about CNNs therefore remains a natural first step when studying deep learning for computer vision. 

Miscellaneous