I was able to finish reading chapter 3 of the MML book and do all the end of chapter exercises in around 3 days. I then moved on to reading chapter 4. I got stuck on trying to figure out singular value decomposition for a while but ended up getting the hang of it after doing some practice questions on it. I am now halfway through the chapter 4 end of chapter exercises. I am pretty sure I will finish them by the morning math session I have scheduled for myself tomorrow and move onto doing calculus. I think I covered alot of the chapter 3 and 4 MML book content when I was reading Paul's math notes, which is why I was able to finish the MML book chapters 3 and 4 so quickly compared to chapter 2.
On another note, I have realized that I absolutely suck ass at reading math proofs when there are no numbers involved. I can't seem to reason with symbols and variable names when learning new math techniques, so everytime I want to learn a new technique I have to find a practice problem online and work through it with numbers. It makes reading through definitions and examples of the MML book feel like cancer. I think I am gonna try to solve this by getting good at math proofs. Proofs are basically math questions with numbers so getting good at them will probably make reading math equations I've never seen before easier and help me reason with math better too. I found this textbook online called Mathematical Proofs: A Transition to Advanced Mathematics, and will be reading through the chapters of it to learn basic proof techniques. I have no idea how I am gonna find the time to read this textbook but I will probably figure it out somehow.
I want to end off this weekly update with a timeline of how my machine learning progress is going. I have spent almost 7 weeks doing linear algebra at this point, which is much longer than I thought I would spend on it. I figure that calculus, statistics, and probability are probably going to take at least up to December to finish also. Probably longer too. I guess learning new things just takes a shit ton of time to do. I still got no idea or intuition on what this math I have learnt will help me do but doing machine learning in the future sounds interesting so I'll just keep learning more math until I can understand the ML stuff.
Thats all for this week's update. See you next week.