What to read on machine learning subject matter?

Question

Books

The question of what to read as an introduction to neural networks has already been raised here, I want to put it from a slightly different angle.

Currently, there are many free and commercial tools for modeling neural networks and other (eg Bayesian networks, SVM, etc.) machine learning methods. The implementation of such a library and its application are two well-separated tasks that require specialists of a completely different “flight”: if a programmer-mathematician is needed to implement the algorithm, then to apply to an applied task, a subject specialist is needed who understands the problem in question and the area to which she applies. It would not be rational to demand from the latter knowledge of the mathematical apparatus of machine learning methods, it is much more constructive to provide him (so that he can choose and use the optimal one) a purely practical description of the available methods (what is at the input (requirements for data preparation), what is the output, methods for choosing the parameters of the architecture and the learning process, types of tasks to be solved, other specifics, a simplified (sufficient for a general idea of ​​the principle of operation) description of the internal structure). What is there to read in this vein (to cover as many different methods as possible)?

Answer 1

If we are talking about getting acquainted with machine learning problems and various methods for solving them, I can recommend the machine learning course from Coursera. But it's not to read, it's to watch

Answer 2

Or Mestetsky's lectures.

Answer 3

There is such a selection of books https://blog.adn.agency/15-knig-po-mashinnomu-obuc...