Resource recommendations for machine learning

[Luqman Saleem]

I am planing to change my field (in PhD) and learn Machine Learning to differentiate different phases of strongly correlated matter. I learned Monte Carlo method in my MS and have intermediate level knowledge of topological insulators.

Before completely getting into Machine Learning, I want to go-through an introductory level book/article of Machine learning for physicists. I want to know if it is too difficult for me to learn. (is it really very difficult?)

Do you know any books/articles in which Machine Learning is explained in context of Physics?

 

[jedishrfu]

Hands On Machine Learning by Aurelian Geron may fit the bill. I’ve looked at several books on ML and while each has some good features this one seems to be the best of the bunch.

https://www.amazon.com/dp/1491962291/?tag=pfamazon01-20

 

[Borg]

@jedishrfu posted this media link on what is a neural network that I found to be the best starting point for understanding machine learning. I've studied a lot of machine learning courses and algorithms over the last year and none of the courses described it the way that that video series did. Without it, I would still be pretty confused in places.

 

[Demystifier]

I like the online lectures by the physicist Michael Nielsen
http://neuralnetworksanddeeplearning.com/especially Chapter 4
http://neuralnetworksanddeeplearning.com/chap4.html

 

[Borg]

I read that and ran the code. Very good online book!