A First Course in Machine Learning

by Simon Rogers, Mark Girolami

A First Course in Machine Learning covers the core mathematical and statistical techniques needed to understand some of the most popular machine learning algorithms. The algorithms presented span the main problem areas within machine learning: classification, clustering and projection. The text gives detailed descriptions and derivations for a small number of algorithms rather than cover many algorithms in less detail.

Requiring minimal mathematical prerequisites, the classroom-tested material in this text offers a concise, accessible introduction to machine learning. It provides students with the knowledge and confidence to explore the machine learning literature and research specific methods in more detail.

Genres: Artificial Intelligence

427 pages, Paperback

First published October 25, 2011

About the author

Simon Rogers is the founding editor of the Guardian’s Datablog and has won numerous awards for his work, including a Royal Statistical Society’s award of excellence in 2012.

Reviews

[Charlie · Rating 5 out of 5]

An excellent introduction to the concepts of machine learning which builds up from first principles. I was fortunate enough to have attended the university course run by Simon Rogers who is an excellent teacher and also a lovely human being. This book requires little mathematical background to start with and introduces all of the necessary linear algebra, calculus, and statistics in a clear and concise way as the book develops and complexity increases.
After learning these ML concepts it's also great as a reference to particular ideas and ML algorithms. However, it is rather light on the actual programming side of things in the text, but there are exercises in maths and coding provided for free alongside to allow you to practice and implement the concepts taught in the book. Highly recommended!

[B R · Rating 3 out of 5]

I think this book might very well have saved me from failing my statistical machine learning course! They introduce all the mathematical concepts you need (results from linear algebra, vector calculus, and the properties of multivariate Gaussians) as the need arises, which makes it far far easier to remember/internalise the relevant mathematics. Additionally there are beautiful explanations of Gaussian Processes and famous sampling algorithms like Metropolis-Hastings. The only downside is that the authors claim that there is python code available for the whole book on their website, but when I went there to take a look, all I saw was yucky R and Matlab code. Still, a very good resource for statistical ML, particularly the Bayesian approach, and for the mathematics and ML concepts needed to dive into deep learning.