These are the six must read books if you are into software and want to build a really good career in artificial intelligence.
1) Artificial Intelligence | Third Edition | By Pearson: A Modern Approach
The book is awesome..total theoretical but it contains each n everything we need to know about AI, from the beginning of AI to its subcategories..everything in depth..very nice book actually..contains all the algorithms that come under AI or depicts AI, i.e. every possible algorithm that we can’t even think of..awesome..loved it..a must have book for all computer science students or scholars.
2) Algorithms to Live By: The Computer Science of Human Decisions
Author talks about real life instances where computer algorithms can be applied. It covers topics like optimal stopping, explore/exploit, caching, scheduling, bayes rule, overfitting, randomness, networking, game theory etc. For people who are computer science professionals this would be a easy read, may not be so for others.
The author should have provided a summary of every rule with the underlying assumptions at the end of each chapters. Its easy to misinterpret the rules if one does not understand the underlying assumptions. For example the ‘37% stopping rule’ for the secretary problem is only valid with he underlying condition that ‘once the candidate is rejected he can’t be recalled’. A simple summary of these rules written as hypothesis at the end of each chapter would have made this book much better.
3) Hands-On Machine Learning with Scikit-Learn, Keras and Tensor Flow: Concepts, Tools and Techniques to Build Intelligent Systems
If you are a practitioner of ML, I’d suggest you to buy this book with eyes closed. The author doesn’t go extremely deep into the theory behind most of the algorithms, because well… that’s not the primary focus of this book. It’s more hands-on, as the title suggests, and I think the contents justify that pretty well. For a more comprehensive treatment to the theory behind the algorithms, I’d suggest to go with ‘The Elements of Statistical Learning’ or ‘Bayesian Reasoning and Machine Learning’. Both are freely available online.
I had the 1st edition of this book but still chose to buy this new and improved iteration mostly because of the following reasons.
1. Addition of new materials (e.g. more unsupervised learning, more deep net techniques, new CNN architectures, etc.)
2. Migration of codebase to Tensorflow 2.0.
3. Discussion on Training and Deployment of Tensorflow models at scale.
This is a colored edition, unlike the 1st edition of the book. It helps immensely, especially for the visualizations and graphs. The page quality is very nice too, and they have a glossy finish.
4) A Thousand Brains: A New Theory of Intelligence
You may find the part 1 boring but it’s very important to build that context. Part 2 talks about limitless opportunities and possibilities (won’t be wrong to say its like prophecy) on existence through preservation of brain. I specially enjoyed the part where the author imagines the ability of downloading the human brain to a computer before leaving his body, in a way keeping his consciousness alive even after death. I enjoyed it!
5) imusti The Elements Of Statistical Learning: Data Mining, Inference, And Prediction, Second Edition
Regarding the book I don’t think it requires any introduction. Anyone who is wanting to learn statistical learning and wanting to learn how various models of machine learning really works should go through this book.
6) An Introduction to Statistical Learning: with Applications in R
I am still going through this book but have gone through enough to write a review. The book is really good one to understand the different class of problems and algorithms that we have with data and the predictions you can make with them. The language of the book is very lucid and helps in having a read through quite easily. The other good part is the algorithms and its concepts are discussed in reasonable detail without delving deep into the mathematical proof of the core formula relating to these algorithms which is very good for people who have lost touch with hard core mathematics so to speak. Another very good part is each chapter has an Lab of sorts where it uses R to show examples of how a particular learning model can be put in place with the data sets. All in all it is a self contained book. So just have to install R and you get going into a very interesting journey with data and its learning algorithms. 5 stars from me for this wonderful effort in compiling this book for the authors.
We are sorry that this post was not useful for you!
Let us improve this post!
Tell us how we can improve this post?