Theoretical resources/books?

Hi all,

I have been into machine learning for a while now. Being a programmer first I feel I have a good foundational knowledge of the practical side of machine learning. I know the tools and methods well.

However I feel that there is a big part of knowledge missing which is the theoretical/mathemathical side to machine learning.

Having always been eager to keep teaching myself I figured I'd delve deeper into the matter.

I have a background in CS and college level math/stat knowledge (been a while since I fully used that knowledge though).

I was wondering if you guys had any good theoretical books, or more preferable, free resources, that I could get to learn the mathemathics behind machine learning.

My main interest goes out to neural networks but other resources are more than welcome! Right now I get how they work, and how some other methods work such as SVM, logistic regression,... However I feel that even though I KNOW how they work, I do not yet fully UNDERSTAND.

I know the difference between supervised/unsupervised learning, I have participated in some kaggle challenges, succesfully using sklearn and other libraries to train and finetune commonly used models.

What I do need: beginner-intermediate mathemathical books/resources on machine learning/neural networks

What I don't need: practical books that teach about using a ML library like sklearn, torch, ...

In other words, I am currently viewing machine learning very much from the programming/practical side of the spectrum, but I would like to get more into the theoretical/mathemathical side.

If the uses programming exercises or the likes, that's not a problem, as long as the main focus is on the theory/mathemathics and the reasons for the theory/math.

I hope that's clear enough and that you guys can help, thanks!