Today, I’m going to take a simple approach to the blog post. I’m working on some other things so, I need to scale this one back. So I’m going to give you 3 books that I recommend to learn general machine learning. I think that these books have been influential to the way that I think about machine learning and they are great resources to have on your book shelf. I promise to come back with an awesome tutorial next week. Since I’m doing a pretty light blog post this week, let’s get down to business and introduce these great books!
And just so that you know, by way of disclosure, if you click on the Amazon links to these books, I will recieve a commission from Amazon. I picked these books, and Amazon only provided the links. Regardless, of the commission, I would still recommend these books to you.
Programming Cognitive Intelligence
This book was my first introduction into programming in python. It was super influential to the way that I think about solving problems with code. It was also a fairly advanced book in terms of actually programming machine learning applications.
Why is it great?
What makes this book so great is that it not only describes machine learning in great detail it goes a step further. It teaches you how to code a machine learning application from the ground up. Technically speaking, you could start off with nothing but the python standard library, work your way through all of the examples in the book, and at the end of it all, you would have a pretty decent, pure python, machine learning library. That’s pretty cool. I think that you should stick with sklearn, but if you want to know what is going on under the hood of these algorithms, there is no other book, in my opinion.
My Thoughts:
You should take some time to read this book. It is really good. It is probably one of the best books that I have ever read on the subject of machine learning. It takes a hands on approach and it will make you a better programmer. It is definitely worth investing the money to have this book as a desktop reference.
Handbook of Statistical Analysis and Data Mining Applications
This was actually one of my text books back in college. It is kind of old now, but it is still a little on the spendy side of the market. It does, however, give you a brief overview of all of the main algorithms used by modern practitioners, aside from deep learning. The book was published just before the recent craze for deep learning really took off. That being said the book is quite good without getting into the weeds of programming and math. It teaches at a very basic level.
Why is it great?
This book is great because it gives a very clean break down of all of the major algorithms used by practitioners today. If you want to understand machine learning on a very high conceptual level, and you don’t necessarily want to dive into the guts of an algorithm, but still want to know what the algorithm is doing, this book is for you. I think it is a great introductory book into machine learning.
My Thoughts:
If you have the money and the desire, then by all means, pick up this book. I wouldn’t sell my copy. However, I also wouldn’t have bought it at all had it not been one of my text books. Also, I do kind of feel like it is an extended brochure for particular software vendors trying to capture the market by providing textbooks to college students so beware of that. I think free tools like R and python do a much better job. However, if you use the tools that they promote, you wouldn’t be doing a bad thing, just realize that you got marketed to, and you bought their product.
Deep Learning
I’ve mentioned this book before here. It is a book that teaches you about Deep Learning. Again this is a graduate level text book, so it is a pretty hefty read, and you need to have some decent math behind you. But if you want to understand what and how these deep neural network models work this is a must read.
Why is it great?
The book is great because it is the only game in town for a topic that is so relatively young. It goes into a lot of detail so that you really understand what is going on under the hood. The deep learning models are super important to lots of different tasks in speech recognition, image recognition, text analysis, and even playing games autonomously, driving cars, and more! This is where all the action is going these days. So we should spend some time learning about what these things are and how they work, right? This book is like the only way to do that!
My Thoughts:
This book is amazing. It has a section that covers more traditional machine learning models, and the authors are fairly quick to point out that these models are not too different than the deep learning models in the sense that the models are just optimizing a cost function. That being said, the book is a bit spendy. So you may want to look at the free online version.