The Best Books for Getting Started with ML

Interested in getting your feet wet with machine learning? These are my recommendations for the best books to get you started. I will do my best to keep this list updated as I come across new resources. Of course, there are also numerous online resources available for free. However, textbooks synthesize all the foundational concepts in one place, which makes them great for those who prefer a more structured approach to learning and are new to the field such that they don’t know what they need to learn.

1. Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron Courville.
2. Machine Learning: A Probabilistic Perspective by Kevin P. Murphy. An encyclopedic treatment of classical machine learning topics. Not always up-to-date with the latest and greatest (see Murphy’s latest books for that), but it is a great resource for understanding the fundamentals. You will also find topics covered here that you won’t find anywhere else.
3. Pattern Recognition and Machine Learning by Christopher M. Bishop. One of the earliest treatments of machine learning, and still one of the best.
4. Gaussian Processes for Machine Learning by Carl Edward Rasmussen and Christopher K. I. Williams. The bible on Gaussian processes.
5. Probabilistic Machine Learning: Advanced Topics by Kevin P. Murphy and Mark Steedman. This book builds upon Murphy’s earlier book and covers newer, more advanced topics in generative modeling.
6. Understanding Machine Learning: From Theory to Algorithms by Shai Shalev-Shwartz and Shai Ben-David. This one is not for the faint-of-heart. Heavily inclined towards theory and covers many topics that are not typically introduced in introductory machine learning courses. However, if you can master this book, you will have a solid foundation in machine learning theory and be ready to begin digging into research in that area.
7. An Introduction to Manifolds by Loring Tu. For those interested in geometric deep learning.