Machine Learning Books You Must Read in 2020

Original article was published on Deep Learning on Medium

Machine Learning Books

In this article, we will explain briefly about some of the best books that can help you understand the concepts of Machine Learning, and guide you in your journey in becoming an expert in this engaging domain. Moreover, these books are a great source of inspiration, filled with ideas and innovations, granted that you are familiar with the fundamentals of programming languages. Read on to know more —

1. Machine Learning for Absolute Beginners: A Plain English Introduction

Author: Oliver Theobald

Publisher — Scatterplot Press

Difficulty Level: Beginner

Get Book hereAmazon

Cover of the book “Machine Learning for Absolute Beginners”

As the title explains, if you’re an absolute beginner to Machine Learning, this book should be your entry point. Requiring little to no coding or mathematical background, all the concepts in the book have been explained very clearly.

Examples are followed by visuals to present the topics in a friendlier manner, for understanding the vitals of ML.

Oliver Theobald has simplified several complex topics related to ML, such as its basics, and other techniques such as Data Scrubbing, Regression Analysis, Clustering, Bias, Artificial Neural Networks, and more in his book. The book also provides additional resources to further learning.

“Analysis As the ‘Hello World’ of the machine”
Oliver Theobald

2. Deep Learning

Author: Ian Goodfellow, Yoshua Bengio and Aaron Courville

Publisher — MIT Press

Difficulty Level: Beginner

Get Book here — Amazon

Cover of the book “Deep Learning”

Regarded as a very beginner-friendly book, it introduces you to a wide range of topics on Deep Learning while also covering related aspects of Machine Learning.

The fundamental concepts of DL are thoroughly explained in this book from scratch, for a stronger foothold in the domain. The book explains relevant concepts of Linear Algebra, Probability and Information Theory, Numerical Computation, industry-standard techniques such as Optimization Algorithms, Convolutional Networks, Computer Vision, and research topics such as Monte Carlo methods, Partition Function. Sufficient supplementary material is bundled for a deeper understanding.

“Written by three experts in the field, Deep Learning is the only comprehensive book on the subject”⁠ — Elon Musk, cofounder and CEO of Tesla and SpaceX

3. Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems (First Edition)

Author: Aurelien Geron

Publisher — O’Reilly Media

Difficulty Level: Beginner

Get Book here Amazon

Cover of the book “Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow”

Easily one of the best-selling books out there for anyone planning to start with Machine Learning or an enthusiast in the domain. Requiring prior knowledge of the Python programming language, it explains some of the most-used ML libraries Scikit-Learn, Keras, and TensorFlow 2, for building intelligent systems.

Intuitively explained concepts and easy to implement examples allow for smoother practical implementation and understanding. Topics included are Support Vector Machines, Random Forests, Neural Nets, Deep Reinforcement Learning, Eager Execution, Time-Series Handling, and more. The book contains updated code examples for several libraries, and APIs involved.

Supplement: You can also find the lectures with slides and exercises on GitHub.

“In Machine Learning this is called overfitting: it means that the model performs well on the training data, but it does not generalize well.”
Aurélien Géron

Is this the best book on Machine learning?

Check out the 2nd edition of the book —

4. Machine Learning (in Python and R) For Dummies

Author: John Paul Mueller and Luca Massaron

Publisher — For Dummies

Difficulty Level: Beginner

Get Book hereAmazon

Cover of the book “Machine Learning (in Python and R) For Dummies”

All books from the famous “Dummies” series have been extremely newbie-friendly. This book, just like others in the series, has its concepts laid out in a manner that readers find easy to follow.

The book includes introductory concepts and theories in ML along with the tools and programming languages involved. The topics covered in the book start with installing R on Windows, Linux and macOS, followed by Matrix Creation, working with Vectors, and Data Frames, working with RStudio or Anaconda to code in either R or Python. It is a handy guide for fundamental concepts of data mining and analysis.

“As a kind of learning, it resembles the methods humans use to figure out that certain objects or events are from the same class, such as by observing the degree of similarity between objects.”
John Paul Mueller

5. Machine Learning in Action

Author: Peter Harrington

Publisher — Manning Publications

Difficulty Level: Beginner

Get Book here Amazon

Cover of the book “Machine Learning in Action”

A valuable book aimed at giving developers a hands-on experience of techniques required for Machine Learning. It’s an equally essential book for familiarizing oneself with ML related Python code snippets, although requiring prior experience with Python.

The book contains code for various algorithms for Statistical Data Processing, Data Analysis, and Data Visualization along with tasks such as Classification, Forecasting, Recommendations, Simplification, and more. With minimal theory, the book cuts straight to the practical implementation of these algorithms.

6. Pattern Recognition and Machine Learning

Author: Christopher M. Bishop

Publisher — Springer

Difficulty Level: Intermediate

Get Book hereAmazon

Github repo:

Cover of the book “Pattern Recognition and Machine Learning”

Directed towards individuals who have a fundamental idea of Pattern Recognition and Machine Learning, this book assumes readers have some degree of prior knowledge in multivariate calculus and algebra.

The concepts in this book aim to explain the recent developments in the underlying algorithms and techniques in the domain of ML. Covering widely used topics such as Bayesian Methods, Regression, Classification, Neural Networks, Graphical Models, Sampling Methods, and more, this book is highly suitable for understanding ML, Statistics, Computer Vision, and Mining. The book comes fully stacked with a broad range of exercises and additional material.

7. An Introduction to Statistical Learning (with applications in R)

Author: Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani

Publisher — Springer

Difficulty Level: Intermediate

Get Book here Amazon

Cover of the book “An Introduction to Statistical Learning (with applications in R)”

This book, although requiring some prior knowledge of linear regression, is an excellent tool for understanding the concepts of Statistical Learning. By providing a balanced insight into how to make use of large and complex datasets, it aims to educate a wide range of statisticians and non-statisticians alike and enable them to understand the data in their hands.

It covers several vital concepts of Statistical Learning, such as Linear Regression, Classification, Tree-Based Models, Support Vector Machines, Resampling Methods, and more. Various examples and tutorials make the learning process more enjoyable, and it includes several R labs, demonstrating the implementation of these statistical methods.

8. Applied Predictive Modeling

Author: Max Kuhn, and Kjell Johnson

Publisher — Springer

Difficulty Level: Intermediate

Get Book here Amazon

Cover of the book ”Applied Predictive Modeling”

Regarded as an exceptional reference book for many of the Predictive Modelling concepts, this book requires a sound understanding of statistics, R programming language, and Machine Learning concepts. The author has focussed on explaining data collection, manipulation, and transformation process as it is often overlooked in ML books.

The applied nature of this book makes it an excellent choice for analyzing real problems faced by industries. Readers can dive into data preprocessing, splitting, and model tuning, followed by regression, classification, handling class imbalance, selecting predictors.

9. Machine Learning for Hackers: Case Studies and Algorithms to Get You Started

Authors: Drew Conway & John Myles

Publisher — O’Reilly Media

Difficulty Level: Intermediate

Get Book here Amazon

Cover of the book ”Machine Learning for Hackers”

As the title says, this book is not for hackers but for people who are interested in the hands-on case studies. Requiring a strong programming background, this book aims to train you with the algorithms driving Machine Learning. Various chapters focus on each of the problems in ML, such as Classification, Optimization, Prediction, and Recommendation.

The book also trains you in R, and how to analyze datasets and gets you started on writing simple ML algorithms. One significant way it differs from other books is its low dependency on maths to teach ML.

10. Programming Collective Intelligence: Building Smart Web 2.0 Applications

Author: Toby Segaran

Publisher — O’Reilly Media

Difficulty Level: Intermediate

Get Book here Amazon

Cover of the book “Programming Collective Intelligence”

Considered by many as the best guide for Machine Learning, this book prefers to teach you the implementation of ML, assuming you know Python. It includes steps for creating algorithms and programs for accessing datasets off websites, collecting data on your own, and analyzing and making use of data.

Taking you into ML and statistics, the book includes examples for crawlers, indexers, optimization, PageRank algorithms, filtering techniques, decision trees. Aimed at walking you through the entire process of creating algorithms at your pace, this book does its job excellently.

11. The Elements of Statistical Learning: Data Mining, Inference, and Prediction

Author: Trevor Hastie, Robert Tibshirani, and Jerome Friedman

Difficulty Level: Expert

Publisher — Springer

Get Book here Amazon

Cover of the book “The Elements of Statistical Learning”

This book focusses on concepts rather than the mathematics behind the concepts. It holds a vast collection of ideas about the implementation of Statistical Learning in several sectors. Filled with relatable examples and visualizations, it should be an essential piece in any statistician or data mining enthusiast’s library.

The book covers supervised and unsupervised learning, including topics such as Support Vector Machines, Classification Trees, Neural Networks, Boosting, Ensemble Methods, Graphical Models, Spectral Clustering, Least Angle Regression, and Path Algorithms, to name a few.

12. Python Machine Learning

Author: Sebastian Raschka, and Vahid Mirjalili

Publisher — Packt

Difficulty Level: Expert

Get Book here Amazon

Cover of the book “Python Machine Learning”

Assuming you already have a strong understanding of many of the core notions of Python and Machine Learning, this book cuts straight to the practical implementation of the concepts. The concepts in the book include up-to-date explanations of NumPy, Scikit-learn, TensorFlow2, and SciPy. The book prepares you to undertake real-world challenges by teaching you from the real-world challenges faced in the industry. It includes various topics such as Dimensionality Reduction, Ensemble Learning, Regression, and Clustering Analysis, Neural Networks, and more.

“Eventually, the performance of a classifier, computational power as well as predictive power, depends heavily on the underlying data that are available for learning. The five main steps that are involved in training a machine learning algorithm can be summarized as follows: Selection of features. Choosing a performance metric. Choosing a classifier and optimization algorithm. Evaluating the performance of the model. Tuning the algorithm.”
Sebastian Raschka, Python Machine Learning