This looks cool: Mathieu Blondel and Vincent Roulet have posted a first draft of their book on arXiv:
https://arxiv.org/abs/2403.14606
@tdietterich I combed through it (in a rush). It's a nice very dense collage of much of the core math of today's ML/DL scene. There's not much of an arc though, and chapters aren't held together by much, so the topic still deserves a much more didactic textbook some day.
@christiankothe I think the authors are looking for feedback. I would like to see more discussion of differentiable programming as a new programming paradigm. But as you say, they mostly dive directly into the details. I'll try to find time to write them a note.
@tdietterich I'm actually impressed that they find time for a book at all amid fiercest competition with OpenAI, and now from a bit of an underdog position. There could also be more on how this can be synthesized into a probabilistic programming language like e.g. numpyro (sampling, variational inference).
