I am a Logistics student. I like the book "Essentials of Economics" by Krugman, Wells and Graddy in that it is concise, easygoing and not only a beginners book (it gradually approaches advanced subjects thus paving the way for further rigorous Economics course) so any one interested in Economics can learn it even if he/she never studied the subject before. Also, I am very interested in AI and Machine Learning and acknowledge their importance in this our postmodern era and I am self learning Real Analysis and web site development. What are some good introductory books on Machine Learning and AI like Krugman, Wells and Graddy's "Essentials of Economics"?
What are some good books on Machine Learning and AI like Krugman, Wells and Graddy's "Essentials of Economics"
$\begingroup$ I thank everyone from the bottom of my heart! I accept all answers as "the Answer"! So I didn't tick any. $\endgroup$– user36339Apr 4, 2019 at 9:05
$\begingroup$ @Anti-AmericanAnti-Zionist let's keep any politics off this SE; consider whether your username is relevant or helpful for a data science site. I've removed your comment above. This Q is borderline closeable as opinion-based but I left it as a wiki as it has gotten some useful responses. $\endgroup$– Sean OwenApr 4, 2019 at 9:36
The two books that come into my mind are:
They both start from the basics and escalate while moving on.
Also thanks for your recommendation, I'll take a look at it because I want to jump to finance at some point in my career:)
What do you want to learn in AI and Machine learning? Artificial Intelligence covers many practical applications, so your question might be a bit vague here. I will suggest you books on Machine learning itself, as it is as a part of Artificial Intelligence.
Simply stated the goal of Machine learning is two-fold: inference and prediction. Inference: the goal here is to understand the relationship between input variables and output variables. If I change the values of the inputs, how do the output values change? prediction: Here we are not as much interested in how the data changes, but just want to know the value of the output variable.
So, in general, you should be interested in Statistics, more specifically concerning prediction and inference. That's it, except that doesn't help you decide which books to purchase.
Here goes the list (it's a popular one)
If you want to learn about machine learning algorithms in a relaxed and fun manner, good if to take up if the next books give you headaches. Certainly worth reading.
An Introduction to Statistical Learning with Applications in R
This book is the most approachable one in the list. It requires some understanding of mathematics to understand certain formulas, but the text is still written in a way that will make concepts clear before you dive into the math. Make sure you do the exercises with R. It's a good skill to pick-up and it will make the theory much more tangible.
This book and next one in the list are freely available online, but if you want you can still purchase paper versions on amazon. I linked you the free versions.
The Elements Of Statistical Learning
This one picks up where ISLR left off. it is more math heavy and explores new concepts. You will find some overlap with the first book which will help solidify the concepts you learned in the first book.
These first three books will already ease you quite into the field. However if you decide to become more serious about learning, the following books should definitely be on your reading list:
Pattern Recognition and Machine Learning
The best advice I can give you with these books is to read them from cover to cover. Don't read too much at once, take breaks and try to explain what you read to yourself. It can often make sense on paper and then not so much when you say it aloud. Don't look at the formulae as something to skip. Instead, look at them like lego blocks. Each symbol has a meaning that is defined in the index at the beginning of each book. Try to explain each symbol in the formula; Then explain how the symbols interact. Once you understand the formula, try to think what happens when certain symbols change values. You'll get a very firm grasp of the formula that way. The field of AI and ML has a lot of jargon it can become overwhelming. By really understanding how certain algorithms work you will stop being fooled by the fancy names and start to realize that there is a lot of repetition.
AI and Machine Learning is a big field. If you want the broadest nontrivial introduction, you should check out:
Machine Learning: A Probabilistic Perspective.
It covers everything from classical statistical methods to graphical models and deep learning. If you are specifically interested in topics having more to do with AI than machine learning, I think you would enjoy learning about reinforcement learning. Possibly as a result of recent renewed interest in the field, a second edition of
Reinforcement Learning: An Introduction
has just come out. The original version was quite good and this new one has, amongst other things, a very interesting section on applications including AlphaGo and Watson (of Jeopardy fame).
If you aren't really sure what you want to study, datasciencetexts.com contains a number of brief descriptions of related and prerequisite subjects, along with recommended books that you might be interested in. (Disclosure: I helped build it.)