joaorico

"I believe one should only read those books which bite and sting. If the book we are reading does not wake us up with a blow to the head, then why read the book? To make us happy, as you write? My God, we would be just as happy if we had no books, and those books that make us happy, we could write ourselves if necessary. But we need the books that affect us like a disaster, that hurts us deeply, like the death of someone we loved more than ourselves, like if we were being driven into forests, away from all people, like a suicide, a book must be the axe for the frozen sea inside us." [2]

It's filled with interactive notes that are very useful for understanding the linguistic and cultural references.

Here's my reading method that I found effective:

  1. Read a section on paper.
  2. Go through the same section on the site.
  3. (Re-)read on paper.

"To understand any subject, above all, a mathematician SHOULD NOT pick up a book and read it.

It is the worst error!

No, a mathematician needs to look in a book, and to read it backwards. Then, he sees the statement of a theorem. And, well, he goes for a walk. And, above all, he does not look at the book.

He says, "How the hell could I prove this?"

He goes for his walk, he takes two hours ... He comes back and he has thought about how he would have proved it. He looks at the book. The proof is 10 pages long. 99% of the proof, pff, doesn't matter.

Tak!, here's the idea!

But this idea, on paper, it looks the same as everything else that is written. But there is a place, where this little thing is written, that will immediately translate in his brain through a complete change of mental image that will make the proof.

So, this is how we operate. Well, at least some of us. Math is not learned in a book, it cannot be read from a book. There is something active about it, tremendously active.

[...]

It's a personal, individual work."

From the preface:

"By Spring 2020, my draft of the second edition had swollen to about 1600 pages, and I was still not done. At this point, 3 major events happened. First, the COVID-19 pandemic struck, so I decided to “pivot” so I could spend most of my time on COVID-19 modeling. Second, MIT Press told me they could not publish a 1600 page book, and that I would need to split it into two volumes. Third, I decided to recruit several colleagues to help me finish the last ∼ 15% of “missing content”. (See acknowledgements below.)

The result is two new books, “Probabilistic Machine Learning: An Introduction”, which you are currently reading, and “Probabilistic Machine Learning: Advanced Topics”, which is the sequel to this book [Mur22].

Together these two books attempt to present a fairly broad coverage of the field of ML c. 2020, using the same unifying lens of probabilistic modeling and Bayesian decision theory that I used in the first book. Most of the content from the first book has been reused, but it is now split fairly evenly between the two new books. In addition, each book has lots of new material, covering some topics from deep learning, but also advances in other parts of the field, such as generative models, variational inference and reinforcement learning. To make the book more self-contained and useful for students, I have also added some more background content, on topics such as optimization and linear algebra, that was omitted from the first book due to lack of space.

Another major change is that nearly all of the software now uses Python instead of Matlab."