This question already has an answer here:
So I’m a first year student and I really want to get better at machine learning. I’ve implemented GANs, Deep Q Learning, autoencoders etc using Tensorflow. I haven’t really played around with linear regression or the more simplistic algorithms.
I quickly realised that while it is fun to be able to implement complex algorithms that have already been created, the novelty begins to die off because I don’t know enough about what’s happening under the hood to be able to optimise correctly, preprocess data correctly, choose the correct loss functions etc. It is clear to me now that I will need to go to the basics to develop this understanding and get sufficient foundations in maths and more basic algorithms.
I am currently churning through “Elementary Linear Algebra by Howard Anton”. I’m only about halfway through, but I can already see how applicable the stuff I’m leaning is to ML. Before I start getting deeper into the actual ML side of things, are there any other concepts that are so fundamental that I should learn them beforehand? What textbooks would be recommended for such topics?
I prefer textbooks over video recordings - just preference. I understand this question is open-ended, but I’d really appreciate some anecdotal answers to shed some insight.