Source: Deep Learning on Medium
My Top 5 Machine Learning Books read this year.
A.I is the new electricity — Andrew N.G.
A.I is growing at a very rapid pace. It is quite hard to keep up with recent updates with more than 100 research papers being published per day. It can be overwhelming for someone who is a beginner , who wants to be a Machine Learning Practitioner. Another reason being with so many online MOOC’s it can be daunting for where to begin! Yes, they do offer valuable education but you don’t have to study everything. Especially, one main question which arises is
‘ Does one need to be good at math for being a Machine Learning Engineer?’
If you have basic understanding of Linear Algebra, Probability, Statistics, Calculus and know how to code in Python it’s a good start. Once you have basic understanding of some of Machine Learning algorithms, choose a project in area you are interested in and build on top of it. If you hit a wall in the process study related concepts and get back to your project.
Even though there are great online resources I find books to be convenient mode to study, revise, for ready reference whenever required. If you are a beginner and want to create interesting solutions with A.I do not fret! I have got your back!
Here’s some of the books which I read this year and found them very much helpful. Books are listed in order keeping in mind that they are beginner friendly.
1. The Hundred page Machine Learning Book
As the name suggests its brief introduction of Machine Learning over 100 pages. It can be read in short amount of time like over a weekend but take your own time and read it at your own pace.
It has a continuously updated wiki with chapters having QR code curated by the author, which you can scan for additional resources. Book is based on read first buy later policy i.e it is hosted for free and you can buy once you find it is good for daily reference.
2. Grokking Deep Learning
Grokking Deep learning is written by Andrew Trask, Sr. Research Scientist at Deep Mind. I am a huge fan of author and have been following his work for over years because he explains every concept with a easy to understand analogy which is based on daily life experiences. In this book you will learn how to build a deep neural network from scratch with Python and Numpy. You can expect a hands-on approach with a strong foundation for building Deep Learning applications.
3. Hands-on Machine Learning with Scikit-Learn and Tensorflow
If you are a beginner and want a good understanding of Machine Learning frameworks this book is perfect choice. It has updated the code examples to use the latest version of Scikit-Learn, NumPy, Pandas, Matplotlib and other libraries. It covers data preparation, data cleaning and data handling methods with in depth explanation of feature engineering and hyper-parameter tuning. It holds your hand throughout from basics of neural network, deep learning to strategies in Reinforcement learning.
Second edition of the book covers few additional chapters considering latest result in deep learning research.
I highly recommend this book if you want a structured road map with hands-on experience with Machine Learning frameworks
4. Generative Deep Learning
What I cannot create, I do not understand.
With rise of GAN’s neural networks are getting better at ability to imagine, paint like artists, compose music and even create human faces. This book offers in depth understanding of Generative Deep Learning and build models step-by-step with practical examples. This book covers key developments in generative modeling from 2014, i.e over past 5 years.
So if you want to know how machines can paint and compose music this is the book you should look for.
5. Hands-On Unsupervised Learning Using Python
“Most of human and animal learning is unsupervised learning. If intelligence was a cake, unsupervised learning would be the cake, supervised learning would be the icing on the cake, and reinforcement learning would be the cherry on the cake. We know how to make the icing and the cherry, but we don’t know how to make the cake. We need to solve the unsupervised learning problem before we can even think of getting to true AI.”
— Yann Lecun (On true AI)
Last but not the least the one area of research which is very interesting is Unsupervised Learning. This book covers basics of Unsupervised Learning and helps you to create end-to-end machine learning project. It covers techniques like clustering, dimensionality reduction etc.
You will learn to identify difficult-to-find patterns in data and gain deeper business insight, detect anomalies, perform automatic feature engineering and selection, and generate synthetic datasets.
Whooooo!!!! Now you’ve everything you need to be a Machine Learning Practitioner. I hope you found this helpful. To Infinity and beyond !!!!!