11
$\begingroup$

I am software/hardware engineer for many years now. However, I know nothing about AI and machine learning. I have a strong background in digital signal processing, and various programming languages (like C, C++ or Swift)

Are there any sources (e.g. books or guides) that teach you AI theory and philosophy right from scratch, and then goes with examples to real life applications, current tools, examples that you can run, etc.?

So, I am not looking for too academic or statistical sources.

$\endgroup$
4
$\begingroup$

If you want a very simple basic book on Neural Networks and not exactly Machine Learning you can try:

These 2 are basic and very simple books which start from scratch and show computations by hand on simple examples. Also these are real life application based books.

If you want to strengthen your theory and learn comprehensively about Machine Learning especially for pattern recognition the best book by far is:

This book requires sound mathematical knowledge especially in the field of Probability Theory, Linear Algebra and Calculus.

Two other very theoretical books on Neural Nets are:

From my experience these are the best introductory books. Also you can check out various OCW run by edx.org like Machine Learning for Data Science and a highly recommended course on coursera.org run by Professor Andrew Ng Machine Learning by Stanford University

I would also suggest you learn Python or R as it is mostly used for Machine Learning due to their powerful scientific packages. Python is very easy to learn and implement programs as compared to C/C++.

Edit: Forgot this book. Although, a little bit advanced some user may find it easy:

$\endgroup$
2
$\begingroup$

You can watch the Machine Learning Tutorial made by Google. Its simple and communication is very clear. In 6 videos you can get a good experience in Machine Learning.

Here: Hello World - Machine Learning Recipes #1

$\endgroup$
2
$\begingroup$

Just take Andrew Ng's (old) Machine Learning class on Coursera, or the Machine Learning class with Sebastian Thrun and Katie Malone on Udacity. Or both. That's a pretty quick way to get a good, solid introduction to the basics of Machine Learning. Then look at the Material from the class at the http://ai.berkeley.edu site and read Artificial Intelligence - A Modern Approach. If you get through that, you'll be well positioned to move on into whatever interests you.

Keep in mind too that you can't completely divorce yourself from the math involved in the field. If you don't already have some background in multi-variable calculus, probability, and linear algebra (mostly matrix operations), then you may need to bone up on that stuff.

$\endgroup$
1
$\begingroup$

You can watch the Machine Learning Tutorial made by Google here: Hello World - Machine Learning Recipes #1. Its simple and communication is very clear.

Six lines of Python is all it takes to write your first machine learning program! In this episode, we'll briefly introduce what machine learning is and why it's important. Then, we'll follow a recipe for supervised learning (a technique to create a classifier from examples) and code it up.

$\endgroup$
  • 1
    $\begingroup$ Welcome to AI, and thanks for contributing. (In future, please try to give a little more detail--this answer was flagged as spam:) $\endgroup$ – DukeZhou Jun 4 '18 at 20:00
0
$\begingroup$

There is an excellent online book that gives a thorough introduction to and training on how to build neural networks is Neural Networks and Deep Learning by Michael Nielson. In the first chapter he uses the example of recognizing handwritten digits and goes over perceptrons, sigmoid neurons, basic neural nets, how to code them in Python, etc. Later chapters go deeper into the basic concepts of neural nets.

I would recommend this book even to those who already have experience with neural networks. It's a great resource.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.