I have about 3 complete days to dedicate to it with little distraction. What is the quickest way to learn Linear Algebra? What online or textbook sources should I use? I just got my hands on the 'No BS Guide to Linear Algebra'.
I want to get the most bang for my buck with my time. Should I go with Khan Academy, Gilbert Strang videos? I have limited time and want to make sure I use it wisely.
Linear Algebra Done Right by Axler
There is a crash course version without any proofs so you won't exactly learn anything linear.axler.net
Since you have to ask this question instead of taking the initiative on your own, I would say the quickest way for >(you), a big guy, to learn Linear Algebra is to go to class, take notes, do the homework exercises, and study for the quizzes and exams.
I am not taking Linear Algebra next semester, I am doing it out of pure interest.
Thanks, I want to see proofs. Is there a version with that included? Is it better to understand Linear Algebra from a computational point, then theoretical, or is it better to learn theoretical then practical or both at the same time?
i'm assuming you're serious and are hoping to get a good base for studying math (e.g. not looking for simple bullshit "linear algebra" for engineers)
Pick up "Linear Algebra - Hoffman & Kunze". Read slowly with pencil and paper, if you don't understand a word or sentence wait and reread and check some small examples until you do. Do every exercise (it's fine if you can't do 1 or 2 per chapter, but only if you try hard to do them and give up after a long while).
The third three chapters will serve you well.
It's always better to read an actual book instead of meme "courses". Axler's and Hoffman's are both great. Pick one of them and read on.
Run through a linear algebra textbook with your programming language of choice and just play around with arrays.
Translate methods for array transformations or problem evaluation into functions instead of importing or using libraries. You can probably get it all done in three days if you're dedicated.
Okay great, thanks.
The full Linear Algebra Done Right is what you want
golibgen.io
Hoffman and Kunze is the standard for rigorous LA but it will take longer than 3 days.
ok thanks i intend to turn off my laptop, phone, etc and grin this out or 3 days straight.
3 days isn't enough though, consider an hour a day or so (or more if you want) for the following few weeks. new information takes time to digest, you don't necessarily need to be working to digest it.
>Catalog
good point. it will be an on-going process but these 3 days will be without distraction
In that case do either book and if you get stuck watch a Strang video about it: ocw.mit.edu
Terence Tao also has good course notes for LA
math.ucla.edu
Shylov, Russian Mathematician also has an awesome intro to LA book that gets difficult quickly.
The full and completely guide to everything LA is of course Artin but it goes into Abstract Algebra too
Cool, thank you guys very much!
Sick, I needed this book very bad.
That's just the abridged version this is the full:
There's a few issues with Axler's book, it's good as a refresher but if you really want to learn full Linear Algebra you go with Hoffman + Kunze, and Halmos's Finite Dimensional Vector Spaces.
Halmos is probably the best math writer I've read that provides brief and to-the-point explanations of the main concepts and their motivations it's all signal no noise. Hoffman & Kunze is written from the point of view of abstract algebra. Rich in examples and exercises, covers more than necessary for any LA course.
After you can go into Advanced Linear Algebra
by Steven Roman which is a grad level text that goes deep into the theory of linear algebra
Youtube vids for me. Look at what worked for you in the past, go from there
>After you can go into Advanced Linear Algebra
I'm sure the poster is knowledgeable and means well, but don't do this op. there's way, way too much to learn with higher priority than specializing in linear algebra, as good as roman's book is
It's req reading to go into advanced ML, machine vision (quaternions), type theory like HoTT, cryptography, signal processing, developing new algorithms in compression ect.
In a seminar on algorithmic problems of group theory complexity, and applications to crypto I attented they left almost every detail as "See this in Roman's Advanced LA book" which is why I picked it up.
well yeah, if you're attending a seminar on algorithmic problems of group theory complexity applied to crypto, you already know some group theory, some algorithms, some ring theory and basic number theory, and I assume you already know other pieces of the basic math curriculum like analysis and some probability since you're doing a lot of probability in ML. OP presumably knows noting but linear algebra after he learns that book
what didn't you like about axler's book? i thought it was pretty good
so which should i pick up? i already started with axler now im confused. yes i am starting with 0 LA
you're good with axler
"Advanced linear algebra" is literally advanced: you need tons of more stuff before that, including a full book in linear algebra and some abstract algebra
OK thanks. I'll stick with Axler. I don't know abstract algebra yet, but will pick that up a bit later. Really been wanting to get my teeth in LA
Watch the linear algebra series from 3blue1brown on Youtube.