“What we want is a machine that can learn from experience“
Alan Turing
There is no doubt that Machine Learning has become one of the most popular career choices nowadays. According to a study, Machine Learning Engineer was voted one of the best jobs in the U.S. in 2019.
Looking at this trend, we have compiled a list of some of the best (and free) machine learning books that will prove helpful for everyone aspiring to build a career in the field.
Enjoy!
Best introductory book to Machine Learning theory. Even paid books are seldom better. A good introduction to the Maths, and also has practice material in R. Cannot praise this book enough.
This free online book is one the best and quickest introductions to Deep Learning out there. Reading it takes only a few days and gives you all the basics about Deep Learning.
It is one of the most famous theoretical Machine Learning books so you don’t need to write much of an intro.
The bible of Deep Learning, this book is an introduction to Deep Learning algorithms and methods which is useful for a beginner and practitioner both.
Really good treatise on Machine Learning theory.
Non Technical product managers and non-machine Learning software engineers entering the field should not miss this tutorial. Very well written (Slightly old and doesn’t cover Deep Learning, but works for all practical purposes).
Wonder how Google thinks about its Machine Learning products? This is a really good tutorial Machine Learning product management.
Monologue covering almost all techniques of Machine Learning. Easier to understand Maths (for people afraid of difficult Mathematical notations).
Monologue covering almost all techniques of Machine Learning. Easier to understand Maths (for people afraid of difficult Mathematical notations).
Machine Learning guide for absolute beginners.
A detailed treatise on Machine Learning mathematical concepts.
Feature Engineering and variable selection are probably the most important human input in traditional machine learning algorithms. (Not that important in Deep Learning methods, but not everything is solved with Deep Learning). This tutorial provides an introduction to different feature engineering methods.
Traditional Machine Learning in recent days has really reduced to running AutoML models (h2o, auto sklearn or tpot, our favorite at ParallelDots) once you are done with feature engineering. (In fact, there are a few methods to do automated non-domain specific automatic feature engineering too). This book covers methods used in AutoML.
A free book that helps you learn Deep Learning using PyTorch. PyTorch is our favorite Deep Learning library at ParallelDots and we recommend it for everyone doing applied research/development in Deep Learning.
Another detailed book on Deep Learning which uses Amazon’s MXNet library to teach Deep Learning.
Francois Chollet is the lead of the Keras Library. His book “Deep Learning in Python” written to teach Deep Learning in Keras is rated very well. The book is not available for free, but all its code is available on Github in the form of notebooks (forming a book with Deep Learning examples) and is a good resource. I read it when I was learning Keras a few years back, a very good resource.
An excellent resource in Bayesian Machine Learning. Uses Microsoft’s Infer.Net library to teach, so you might have to install IronPython to read/implement the book’s examples.
Another book detailing various Bayesian Methods in Machine Learning.
Natural Language Processing is the most popular use of Machine Learning. These notes from a GATech course provide a really good overview of how Machine Learning is used to interpret human language.
The bible of Reinforcement Learning. This is a must-read for anyone getting into the field of Reinforcement learning.
Teaches using Bayesian Optimization and Gaussian Processes for Machine Learning. With variational inference based libraries like Edward/GpyTorch/BOTorch etc., this method is making a comeback.
Going for an interview for a Machine Learning job? These questions might be of help to figure out strategy while answering Machine Learning systems problems.
This book deals with the parts of Machine Learning which deal with computational algorithms and numerical methods to solve like factorization models, dictionary learning and Gaussian Models.
With causality making inroads into Data Science fields, Machine Learning is not free from the discussion too. While no detailed material is available around this, here is a short tutorial trying to explain key concepts of Causality for Machine Learning.
Found the blog useful? Read our other blog to learn all about the best books to help you excel as a data scientist.
