Author: Andriy Burkov
Publisher: Self Published
Publication Date: January 13, 2019
Product Link
Prerequisites: Computer science fundamentals and some college-level mathematics
About This Book
This is an introductory book into machine learning but is easier to read if you already have most of a computer science or mathematics degree. It gives you the fundamentals of everything related to machine learning, enough to give you the required foundation, but doesn’t always go into too many details about any specific topic. As quoted by Peter Norvig on the back cover, “this is the first 100 pages you will read, not the last”.
Who Is This For?
This book is intended for someone new to machine learning, as stated in the preface. That doesn’t mean that they don’t need any kind of background to understand it. To get the most out of this book, I highly recommend a good programming background and have taken at least a few college-level mathematics courses, especially linear algebra. Someone with practical experience in machine learning can also find this book useful in rounding out their knowledge.
Why Was This Written?
The point of this book is to introduce the topic of machine learning in a set number of pages while covering as many topics as possible. It succeeds in that, but there was a great deal left out that would have been equally useful. As such, why not go with a different book? I feel that this book is still a good introduction and will help you understand a more complete book better. In the end, you’ll spend less time and effort learning the core of machine learning.
Organization
This book does a good job starting with the basics and introducing more difficult topics afterward. There is no standard microstructure within the chapters, which does give the text the flexibility to talk about each topic in a way that makes more sense rather than forcing it into a rigid format. On the other hand, you get more information about some topics than others, which can give a lopsided discourse.
Did This Book Succeed?
The aim of this book is to give a complete introduction to machine learning to someone reading about it for the first time before moving onto another, deeper book. I think it does that well enough to be worth reading. It’s a low cost, low time investment that can really pay off in the long run. It simply isn’t meant to be used alone.