Understanding Machine Learning: From Theory to Algorithms

This book focuses to generate a basic understanding of machine learning, and the algorithmic paradigms it offers, in a principled way. Each concept of ML is divided into a block and explained with real-world examples.

Understanding Machine Learning: From Theory to Algorithms, provides a theoretical account of the fundamentals underlying machine learning and the mathematical derivations that transform these principles into practical algorithms.

Following a presentation, the book covers a wide array of central topics unaddressed by previous textbooks.

These include a discussion of the computational complexity of learning and the concepts of convexity and stability; important algorithmic paradigms including stochastic gradient descent, neural networks, and structured output learning.

It also covers emerging theoretical concepts such as the PAC-Bayes approach & compression-based bounds.

Designed for advanced undergraduates or beginning graduates, the text makes the fundamentals and algorithms of machine learning accessible to students and non-expert readers in statistics, computer science, mathematics, and engineering.

The books start with a handy, concise and clear introduction to statistical machine learning and then consistently connect those concepts to the main ML algorithms and when they put into application.

There is a lot of mathematical rigor, which might not be needed if you are not into research. The fundamentals are covered in a better manner, each chapter is about 10 pages long.

A brief summary at the beginning of each chapter gives a clear sense of what will be accomplished in it, and attention to notation makes sure that mathematics supports understanding rather than getting in the way.

Understanding Machine Learning: From Theory to Algorithms, it is definitely not a “how-to” book, but rather a “what & why” book, focused on understanding principles and connections between them. I read the book cover to cover, and I was left with a sense of machine learning as a coherent discipline and a solid feel for the main concepts.

If we talk about writing style, then this book is very well written, it goes so much into detail but it’s still very intuitive. Small chapters are very informative and keep you interested in the topics.

The only thing that I didn’t like about it is that there are 31 chapters in 360 pages, along with exercises but without answers which might go tough for beginners.

Table of Contents

• Introduction