Machine Learning has granted incredible power to humans. The power to run tasks in an automated manner, the power to make our lives comfortable, the power to improve things continuously by studying decisions at a large scale, and the power to create species who think better than humans. This list can go on and on.
Still sceptical about AI and ML? Read what Google's CEO Mr. Sundar Pichai had to say all the way back in 2015:
"Machine learning is a core, transformative way by which we're rethinking everything we're doing. We're thoughtfully applying it across all our products, be it search, ads, YouTube, or Play. We're in the early days, but you'll see us in a systematic way think about how we can apply machine learning to all these areas." - Sundar Pichai, CEO, Google
Those who know of these advancements are keen to master this concept, including us at Analytics Vidhya. When we started with this mission, we found various forms of digitized study material. They seemed promising and comprehensive, yet lacked a perspective. Our curiosity didn't let us rest for long and we resorted to books.
When Elon Musk, one of the busiest men on the planet, was asked about his secret of success, he replied, I used to read books. A LOT'. Later, Kimbal Musk, Elon's brother, said, √??√?¬Ę??He would even complete two books in a day'.
Following are some of the books on Machine Learning and Artificial Intelligence which should be read by everyone.These books are in no particular rank or order.
Machine Learning Yearning:
We had to start with a book by the great Andrew Ng. It is still a work in progress. This book will help the reader get up to speed with building AI systems. It will effectively teach you how to make the various decisions required with organizing a machine learning project. There's no better person to start off this list, in our opinion.
Programming Collective Intelligence:
Programming Collective Intelligence, PCI as it is popularly known, is one of the best books to start learning machine learning. If there is one book to choose on machine learning it is this one. I haven't met a data scientist yet who has read this book and does not recommend to keep it on your bookshelf. A lot of them have re-read this book multiple times.
The book was written long before data science and machine learning acquired the cult status they have today but the topics and chapters are entirely relevant even today! Some of the topics covered in the book are collaborative filtering techniques, search engine features, Bayesian filtering and Support vector machines.
Machine Learning for Hackers:
This book is written by Drew Conway and John Myles White. It is majorly based on data analysis in R. This book is best suited for beginners having a basic knowledge and grasp of R. It covers the use of advanced R in data wrangling. It has interesting case studies which will help you to understand the importance of using machine learning algorithms.
Machine Learning by Tom M Mitchell:
After you've read the above books, you are good to dive into the world of machine learning. And this is a great introductory book to start your journey. It provides a nice overview of ML theorems with pseudocode summaries of their algorithms. Apart from case studies, Tom has used basic examples to help you understand these algorithms easily.
Most of the experts you ask in this field never fail to mention this book which helped them at the start of their careers. It's such a well-written and explained book that we feel it should be made mandatory in every machine learning course.
The Elements of Statistical Learning:
This is quite a popular book. It was written by Trevor Hastie, Robert Tibshirani and Jerome Friedman. This book aptly explains various machine learning algorithms mathematically from a statistical perspective. It provides a powerful world created by statistics and machine learning. This books lays emphasis on mathematical derivations to define the underlying logic behind an algorithm. Keep in mind that you need to have a rudimentary understanding of linear algebra before picking this up.
There's a beginner friendly version of these concepts in a book by some of the same authors, called Introduction to Statistical Learning'. Make sure you check that out if this one is too complex for you right now.
Learning from Data:
This book is written by Yaser Abu Mostafa, Malik Magdon-Ismail and Hsuan-Tien Lin. It provides a perfect introduction to machine learning. This book prepares you to understand complex areas of machine learning. Yaser, a very popular and brilliant professor, has provided 'to the point' explanations instead of lengthy and go-around explanations.
Pattern Recognition and Machine Learning:
This book is written by Christopher M Bishop. This book serves as a excellent reference for students keen to understand the use of statistical techniques in machine learning and pattern recognition. This books assumes the knowledge of linear algebra and multivariate calculus. It provides a comprehensive introduction to statistical pattern recognition techniques using practice exercises.
Natural Language Processing with Python:
Folks interested in getting into Natural Language Processing (NLP) should read this book. It's written in a lucid and clear manner with extremely well-presented codes in Python. Readers are given access to well-annotated datasets to analyse and deal with unstructured data, linguistic structure in text, among other NLP things.
- Artificial Intelligence Books
A Modern Approach: Who better to learn AI from than the great Peter Norvig? You have to take a course from Norvig to understand his style of teaching. But once you do, you will remember it for a long, long time.
This book is written by Stuart Russell and Peter Norvig. It is best suited for people new to A.I. More than just providing an overview of artificial intelligence, this book thoroughly covers subjects from search algorithms and reducing problems to search problems, working with logic, planning, and more advanced topics in AI, such as reasoning with partial observability, machine learning and language processing. Make it the first book on A.I in your book shelf.
Artificial Intelligence for Humans:
This book is written by Jeff Heaton. It teaches basic artificial intelligence algorithms such as dimensionality, distance metrics, clustering, error calculation, hill climbing, Nelder Mead, and linear regression. It explains these algorithms using interesting examples and cases. Needless to say, this book requires good commands over mathematics. Otherwise, you'll have tough time deciphering the equations.
Paradigm of Artificial Intelligence Programming:
This book teaches advanced common lisp techniques to build major A.I systems. It delves deep into the practical aspects of A.I and teaches its readers the method to build and debug robust practical programs. It also demonstrates superior programming style and essential AI concepts. I'd recommend reading this book, if you are serious about a career in A.I specially.
Artificial Intelligence, A new Synthesis:
This book is written by Nils J Nilsson. After reading the above 3 books, you'd like something which could challenge your mind. Here's what you are looking for. This books covers topics such as Neural networks, genetic programming, computer vision, heuristic search, knowledge representation and reasoning, Bayes networks and explains them with great ease. I wouldn't recommend this book for a beginner. However, it's a must read for advanced level user.
Machine Learning Yearning:
Nick Bostrom has authored (or co-authored) over 200 publications, including this book called Superintelligence. Most of the world is enthralled and captivated by what AI can do and it's potential to change the world.
But how many of us stop to think about how AI will affect our society? Are we considering the human aspect at all when building AI products and services? If not, we really should. In this thought-provoking book, Nick Bostrom lays down a future scenario where machines reach the superintelligent stage and deliberately or accidentally lead to the extinction of humans.
This might sound like a sci-fi movie plot, but the way Mr. Bostrom has laid down his arguments and the thinking behind them will definitely sway you and make you take him seriously. We consider this a must-read for everyone working in the AI space.
The Singularity is Near:
Similar to the above idea propounded by Nick Bostrom, Ray Kurzweil's Singularity is Near' delves into the thick depths of superintelligent machines. It is a slightly long read, but well worth it in the end. The way Mr. Ray has described the Singularity is breathtaking and will make you stop in your tracks.
Singularity, as Ray Kurzweil has described it, is the point where humans and the intelligence of machines will merge. Once this happens, machines will be far more intelligent than all of the human species combined. It's not science fiction but a truly poignant description of what might happen in the future if we aren't careful with what and how we work with AI.
Some of the basic questions this book asks (and answers) are (taken from Amazon's summary): How can we grow our prosperity through automation, without leaving people lacking income or purpose? How can we ensure that future AI systems do what we want without crashing, malfunctioning or getting hacked? Should we fear an arms race in lethal autonomous weapons? Will AI help life flourish as never before, or will machines eventually outsmart us at all tasks, and even, perhaps, replace us altogether?
The Master Algorithm:
This is one of our favorite books in this list. Can there be just one algorithm that deals with all the aspects of technology? Instead of building AI products for specific functions, can we build one single algorithm for all functions? This thought is quite similar to what Albert Einstein spent the latter years of his life trying to discover. Pedro Domingos is a masterful writer, and he deals with the intricacies of his subject extremely well.
Books are a wonderful source of knowledge for anyone willing to learn from them. This collection spans various aspects of AI and ML from the mathematics and statistics side to the intangible factors like ethics and impact of society. All of these should be considered together when working on an AI and ML project.