Ask HN: What deep learning papers should I implement to learn?

Andrew Ng has often stated that the best approach (that he has seen) to mastering DL is to start reading papers and then to implement them.

But the question remains where to start.

Now my goal is to curate a list of papers and their difficulty to implement them so that anyone can have a roadmap of papers to learn deep learning.

If you want to help, please rank the papers you think are worth implementing in the following categories: [Beginner] [Intermediate] [Advanced] [Pro]