Any good books on artificial intelligence for self-studying?

[Anti Hydrogen]

Summary:: what are some good books they recommend on the topic?

thanks in advance

 

[sysprog]

What set of aspects of AI are you seeking to learn (or learn about)? Have you studied programming paradigms, discrete mathematics, first-order logic, etc.? What more can you tell us regarding your AI learning goals?

 

[bigfooted]

I don't know what is considered 'the bible of AI/ML/DL' but if you want to do some self-studying with hands-on experience the best thing I think is to learn python (if you haven't done so already) and start implementing neural network algorithms in it... assuming you have already some background in mathematics and programming. There's plenty of accessible books on python and neural networks that do not require a BSc in computer science to get started.

 

[Borg]

Here is my default suggestion list.

Siraj Ravel:
Siraj rocks! Highly informative and energetic videos on everything from python, to ML to cryptocurrencies. Nearly every video has the code on github that you can try out for yourself.

ThreeBlueOneBrown:
At their heart, neural networks are pretty simple. Unfortunately, most tutorials that try to explain how they work are either boringly dry or very limited in their attempts to describe them. If you're just getting started, I highly suggest watching the series on Neural networks before you study anything else related to machine learning.

ThreeBlueOneBrown has lots of other well-produced video playlists covering topics such as Linear Algebra, Calculus, and more. If you want a refresher in difficult subjects that is explained well and leaves you feeling that you actually learned a complex topic, this is the place to go.

Neural Networks and Deep Learning by Michael Nielsen:
I saw a link to this free, online book while watching the ThreeBlueOneBrown neural networks videos. This book takes a deeper dive into the guts of training a Convolutional Neural Network on the MNIST dataset. By following along with the coding examples, you can write your own CNN that achieves over 99% accuracy.

Deep Learning by Ian Goodfellow and Yoshua Bengio and Aaron Courville:
This free, online book is considered the bible of machine learning.

Other sites with lots of examples below. Note that I have see medium.com pages disappear at the whim of the person who created them. So, if there's one there that you like, copy it.
medium.com
towardsdatascience.com
Adam Geitgey has a lot of good tutorials on machine learning.

 

[discoversci]

Statistics book will be just fine. Everything else is available online. AI is just a buzz, nothing more.

 

[lomidrevo]

As already pointed out, it would be helpful if you specify what is your goal, AI is quite a broad topic.

If you want to get a strong theoretical background of ML/DL techniques, then I agree with @Borg: Deep Learning by Goodfellow,Bengio and Courville is very good to start with.

On the other hand, if you want to begin to produce some ML/DL applications as soon as possible, and just superficial understanding of the theory is sufficient to you, I would recommend this https://www.amazon.com/dp/B07XGF2G87/?tag=pfamazon01-20by Aurelien Geron (I've got the 1st edition, not including the Keras API). However, you need to have some experience with programming in Python, especially with numpy and pandas.

 

[lomidrevo]

Statistics book will be just fine. Everything else is available online. AI is just a buzz, nothing more.

I don't agree with you. Yes, statistics is in the core of many of AI techniques, but you cannot say "AI = Statistics" (as you try to suggest).
ML/DL is more complex as it integrates understanding coming from many STEM domains, beside statistics and probability theory, it is mainly mathematical optimization, linear algebra, calculus, information theory etc... Although the current development seems to be pretty independent, Artificial Neural Networks have been influenced/inspired also by neuroscience.