Probabilistic machine learning. Introduction

This classical work contains a thorough modern introduction into machine learning (including deep training), considered through the unifying prism of probabilistic modeling and the Bayesian decision theory. The basic mathematical apparatus is included (including elements of linear algebra and optimization theory), the basis of learning with the teacher (including linear and logistics regression and deep neural networks), as well as more complex topics (including training and training without training without teachers).
Exercises at the end of the chapters will help readers apply the acquired knowledge, and the application has a summary of the designations used.
The publication was based on the book of Cavin Murphy "Machine Learning: A Probabilistic Perspective". However, this is a completely new work that reflects many achievements that have happened in this area over the past 10 years