Self Study roadmap for Machine Learning

[Oats]

Hello, I am a second year PhD student in mathematics currently studying logic. I have recently begun questioning staying in the field, and I really think I may want to work in Machine Learning, at the very least in the theory of it. I am willing to self study from machine learning books, but quite frankly, I don’t know which ones to read, which are good, etc. Can anyone provide a nice roadmap for someone to learn machine learning (and neural networks) with my background? I know a tad bit of programming in Python. If there are any additional math subjects needed for background as well, I can study that too if necessary. Thanks in advance.

 

[jedishrfu]

There are several blogs that you can follow to learn what's hot, what's not, and how to get into the fray. Be aware that there is a lot of hype going on right now and a lot of folks joining the party.

3blue1brown YouTube has a sequence of videos on Neural Nets, Linear Algebra and Calculus that are quite cool.

http://www.3blue1brown.com/

and there's a ton of courses but your best bet is to find someone to recommend one to you:

https://www.google.com/search?q=lea....chrome-ntp-vasco..0.1.15.4...103.ndQBquhc84E

Quantitative Economics has a course in using either Python or Julia for machine learning from a QE perspective:

https://lectures.quantecon.org/jl/index.html

I'm currently playing with Julia as I feel it will take the ML field by storm as it matures. Juli Computing has JuliaPro and JuliaBox products that provide IDE experiences in the language. JuliaPro is akin to Eclipse or Netbeans IDE and JuliaBox is Jupyter Notebooks IDE (which is really cool and great for learning). Both are free at some level.

www.juliacomputing.com

 

[Zap]

Data Science is HOT right now!

Got me learning Python and ish.

 

[Felipe Lincoln]

If you want to get some practical skills in this field I recommend you the Kaggle plataform, they have a section for learning : www.kaggle.com/learn/overview
There you can take part of competitions and get real skills (and even money) in machine learning.

 

[atyy]

Machine learning as a branch of statistics
https://www.amazon.com/dp/0387310738/?tag=pfamazon01-20
Pattern Recognition and Machine Learning by Christopher Bishop
https://www.amazon.com/dp/0262018020/?tag=pfamazon01-20
Machine Learning by Kevin Murphy
https://www.deeplearningbook.org/
Deep Learning by Ian Goodfellow, Yoshua Bengio and Aaron Courville

Machine learning as a branch of control theory
https://www.crcpress.com/Reinforcem...abuska-De-Schutter-Ernst/p/book/9781439821084
Reinforcement Learning and Dynamic Programming Using Function Approximators by Lucian Busoniu, Robert Babuska, Bart De Schutter, and Damien Ernst
http://incompleteideas.net/book/the-book-2nd.html
Reinforcement Learning by Richard Sutton and Andrew Barto