So I'd like to get into machine learning, but I haven't learnt linear algebra or calculus yet. What books would BOT recommend in these topics to a beginner like me?
So I'd like to get into machine learning, but I haven't learnt linear algebra or calculus yet. What books would BOT recommend in these topics to a beginner like me?
>Total price: $240
Did libgen not exist in 2011?
not everything is on libgen
Thanks. And what are some books that are specifically about ML?
Deep Learning Architectures - A Mathematical Approach
Thanks anon, I appreciate it.
>4 calculus books
>Using books for learning in 2023
Go back to school, boomer.
kek, see ya in the next "I sent out 200 CVs, still no replies, it's been 6 months, help" thread
Any information you could possibly want to know about this subject can be found online. If you know where to look for information, it will be more reliable than learning it from a book and it will be more up-to-date than learning it from a book. The sources for the information in such books will be online.
Go ahead and continue projecting your own personal failures on others though...
You don't need much algebra or neuroscience to understand T. Mitchell's Machine Learning. He's the granddaddy, start there.
I miss those days when I didn't read and thought the shallow summaries you can find online were the only things that counted.
You're not going to find detailed, helpful, well-written information arranged in a logically incremental way through a Google search. This isn't about "knowing where to look", unless you're referring to journal sites that still, nonetheless, are worthless to the beginner. The internet is as wide as as a sea and as deep as a puddle when it comes to quality information for pretty much every subject, if the sea just so happened to be chock-full of scammers trying to sell you their terrible and unproven methods on every shore.
Stop being allergic to books and thinking you can get away with the practical application of your shallow understanding, you're only crippling your potential. I bet everyone around you must think you're both incredibly smart but also a complete retard.
>I miss those days when I didn't read and thought the shallow summaries you can find online were the only things that counted.
This says more about your inability to find cutting-edge sources of information than it does about anything else.
>You're not going to find detailed, helpful, well-written information arranged in a logically incremental way through a Google search.
Speak for yourself. If you are incapable of locating the peer reviewed journals where the latest advancements and theories are published, that is on you.
> worthless to the beginner
Tell me you are a retard without telling me you are a retard. I'll bet you probably stop reading a body of text as soon as you encounter a term or concept that you don't understand. Please proceed to filter yourself.
>I barely understand the new and know nothing of what came before
The consoomer of information. Amazing.
Have fun reading the already-outdated information from your books which were compiled from information gathered online.
I only read books written by the geniuses in their fields explaining their topics of expertise. I'm not talking about modern textbooks, those are as bad as the internet.
First, start by realizing every field has remained in stasis for the last 100 years except for maybe neuroscience (which needs a breakthrough, as it is right now it's worthless).
Theories still applied today aren't "outdated". Generally nothing written by geniuses ever is.
Now keep reading journal slop pumped out for tenure just because it isn't "outdated". lol retard
Enjoy failing to understand the cutting edge knowledge because you don't have the background to follow the discussions it comes from.
If the op is "getting into machine learning" without knowing elementary maths he's not after anything that will ever be outdated anyway.
I too prefer 10 minute ESL youtube tutorials
Hands On Machine Learning is pretty good. Unfortunately it uses Tensorflow, but the knowledge is universal and you'll be able to learn PyTorch easily
Are you interested in theory or practical application? Theory will require a lot of reading, which is why you see so many books being posted in this thread. But for practical applications, just one book (e.g. ) is sufficient to get started.
Mostly theory, but practical application seems interesting as well. I definitely going to check out all these books.
I'd strongly recommend starting with Introduction to Statistical Learning, at least the first few chapters. A lot of the meme learning content out there is very focused on what you can do and not very focused on whether it's actually informative and a little statistics context can go a long way to keeping you grounded. The code examples are all in R but I'm pretty sure there's python versions all over the place.
If you are serious about ML theory, you need linear algebra and calculus---probability theory and statistics as well. If you really want to be thorough, you can use
, but going quickly through an introductory textbook in each of these topics is sufficient. Then, the classic introductory texts are
and Elements of Statistical Learning.
If you're not that serious or just want a quickstart, just learn basic single-variable differential calculus, get comfortable with vectors and matrices, and you can make do with a simpler book like
- you'll understand how different models work under the hood.
>haven't learnt linear algebra or calculus yet.
you really want to compete with people who already know this shit?
fast.ai then, kaggle competitions and read anything you don't know
Just pair up khan academy calculus and LA exercises with online info on how to solve them. If you want to go in-depth, grab books or watch classes of course
A probabilistic theory of pattern recognition
If you don't understand what pic related is, it never even began for you.
Pattern Recognition and Machine Learning from Bishop
Book was bought by MS and is now freely available.
https://www.microsoft.com/en-us/research/uploads/prod/2006/01/Bishop-Pattern-Recognition-and-Machine-Learning-2006.pdf