How to get into AI + Facebook AI reading list (the best resources)
HTML-код
- Опубликовано: 11 июн 2020
- Follow me on twitter for my journey as a startup founder! / acecreamu
Hi, I'm Oleksii and I work at Facebook AI Research. The list of the best resources to learn ML and AI is right here:
========== Stage 1 ==========
-Theory:
Linear Algebra, Statistics, Theory of Probability (any courses/books will work)
-Practice:
Basic Python skills, NumPy
========== Stage 2 ==========
-Theory:
Online courses / books on ML and Deep Learning;
Machine Learning course by Andrew Ng, Coursera;
Deep Learning Book, Ian Goodfellow;
Stanford CS231n course;
Recent MIT course: introtodeeplearning.com/;
www.fast.ai/
-Practice:
Pandas, PyTorch, Tensorflow and "official" tutorials (form their websites)
========== Stage 3 ==========
-Theory:
"Classic" papers, e.g. AlexNet, ResNet, BERT;
Reading lists:
deeplearning.net/reading-list/
github.com/ujjwalkarn/Machine...
github.com/floodsung/Deep-Lea...
MILA reading list: docs.google.com/document/d/1I...
Andrej Karpathy's blog: karpathy.github.io/
-Practice:
Random tutorials on the internet, Advanced tutorials, replicating papers
========== Stage 4 ==========
-Theory:
New paper on your topic, conferences, ArXiv, mailing lists and news channels;
distill.pub/;
Sebastian Rudder: newsletter.ruder.io/ ;
THE BATCH: www.deeplearning.ai/thebatch/
-Practice:
Working on toy-projects;
Contributing to open-source projects;
Getting an internship;
Trying to get hands-on experience in any way possible
========= Never Stop =========
These all will be usefull for anyone, engineer or researcher. Also try to concentrate on one particular topic and get an expert in it first, it will be much easier to accomplish and will open you opportunities right away. Don't try to become an expert in everything and also do not neglect "not prestigious" experience. Any experience is very valuable.
#================================
Music used:
YØNAH - California Nights
My homepage: acecreamu.github.io/
My LinkdIn: / sidorovoleksii
My Instagram: / acecreamu
The oblation for youtube gods, please ignore:
machine learning, artificial intelligence, data science, research, faang, facebook, google, apple, amazon, microsoft, mle, ml engineer, ai engineer, ai researcher, get started, courses, resources
#MachineLearning #AI #HowTo - Наука
Let's collect here all the other great resources I forgot to mention.
Let me start: Hugging Face 's blog for NLP
University of Waterloo lectures
Deep Mind & UC Berkeley's RL lectures and assignments
ML Cookbook ml-book.com
EECS 498 of university of Michigan by Justin Johnson ( updated version of cs231n )
Brotha pl tell how much time it should take , i want to do it as efficiently ap👍
Thanks for breaking it down into theory and practice! Really appreciate the links and examples
Amanda Piyapanee Glad you like it
Hello Oleksii, here are some additional resources:
- - - - - - - - - - - - - Basic Topics - - - - - - - - - - - - - - -
-Linear Algebra: The Video Lectures of Prof Gilbert Strang from MIT Opencourseware (Intro to Linear Algebra), there is also a book: Intro to linear Algerba 5th Edition.
- Probability: an awsome book from John A Rice: Mathematical Statistics and Data Analysis, 3rd Edition.
- Calculus: The paper of jeremy howard titled: The Matrix Calculus You need For Deep Learning:
arxiv.org/abs/1802.01528
- On Deep learning: The video lectures from NYU university made by Yann Lecun, Alfredo Canziani and Mark Goldstein.
- On Optimization: Course notes on Optimization for Machine Learning from Gabriel Peyré:
mathematical-tours.github.io/book-sources/optim-ml/OptimML.pdf
---------------------- Advanced Topics----------------------
- The website: papers with code: paperswithcode.com/ for reading and replicating papers.
- The RUclips Channel of Yannick Kilcher: ruclips.net/channel/UCZHmQk67mSJgfCCTn7xBfew.
Great, thank you. "Papers with code" is really good one, I agree.
Thanks a lot for such a great video!You are a gem! :)
Good video with helpful recommendations.Thank you.!!
You are welcome
Great Sir! Really nice and helpful video. The resources you mentioned are very useful.
Would you please make a video on how to learn from research papers and also what ML projects are useful to add to the resume??
Noted ;)
Really super helpful
Helpful content Oleksii
Sumanth Nandamuri great!
CMUs deep learning course is next level. Its worth to mention.
Can u share the link? Thanks
Thank's a lot for this video! I've discovered some new resources for myself. If there is a part 2 of this video)... you may also take a look into the University of Waterloo lectures (youtube also) with very strong math background behind models(if u like to prove math theorems ofc:)) )
Леонид Костюшко I think very few people like this part, haha
Nice recommendations!!
Glad you like it, feel free to add your recommendations for people who will read comments
Can't believe I waited so long for this. Can you pls make a video on how to get a job as a machine learning engineer in facebook discussing the various requirements and skills. That would be great. Keep up the great work brother.
Syed Faizan Ahmed The previous video is about that
Yes I have watched that but don't get me wrong you discussed about how to get into FANG companies as a SWE and little about UX design focusing on to get a referral and all that. What i am saying is that if you don't mind could you do a detailed video on how to get hired as a machine learning engineer in Facebook focusing on what languages to learn, projects and various skills like how to ace coding interviews etc. If you could interview a machine learning engineer from Facebook I would be overwhelmed. Thanks and take care brother.
Syed Faizan Ahmed Got you, will take a note!
@@OleksiiSidorov Thanks brother. Keenly waiting for a video on it.
Thank you
Hi @Oleksii Sidorov
Do you think that it is worth to go through DL specialization? or can I start CS231n just after Machine Learning course by Andrew N?
you can start straight, it's for everyone
Thanks oleskii your talk is always simple and full of content, which makes it more interesting. Looking to be in touch on LinkedIn and share my story as well. Thanks
Nikhil Verma Glad you liked it, sure, let's connect.
Nikhil Verma I would encourage you to share it here better, so other people can help you as well.
Please make a video on that topic
How to get internship(online from anywhere in the world at big tech companies) as Mobile App developer after learning React Native, make complete video.If you'll do it'll super helpful because there is no video on RUclips according to this title that I've mentioned in above.I'm waiting.
Hi Oleksii! I've noticed your resume is 2-page. Do you suggest a one-page/ two-page resume for FB AI residency? Thank you!
It doesn't matter, concentrate on the content and not the form.
Hi ,do u think the pandemic will have a huge impact in hiring this year?
Are u working from home?
And can u please make a video sharing some ML interview questions they asked u ,or u can jus reply here
Yeah, it will have some impact, also hire will continue, just less intense.
We all are working from home and probably companies will keep it like that to 2021 at least
isn't there a threshold for each stage?
This is a really nice video! Thanks a lot. I'd like to know - which ones to decide -> NLP or CV or RL or Audio? Which one to start as a fresher who completed 2 stages and trying into dive into the third stage?
Hi there, I'd like to ask for the eligibility criteria and the financial support, as an undergrad and as a foreigner.
- Is it possible to join the Facebook AI Residency program right after completing the undergraduate program? (with enough of 3 to 4 research papers and internships)
- And how much and based on what criteria financial support is available for the foreigners,
It would be really helpful for me if I get answered.
Thanks for your time.
With kind regards.
Wolf Rage 1. Yes, it's possible. 2. Everyone gets full financial coverage regardless of citizenship, you shouldn't worry about that.
@@OleksiiSidorov Thanks again, 🥰🥰.
Looks like some people are giving false info in my college. Anyways, thanks a ton for your time.
so where were your level when you joined in FB as AI resident?
I had some papers published already and have completed some research projects
Ngl but if you showcase a bit other DL/AI as well like about your life in general( u still work out?) would be great like time management and stuff . People usually have a notion that researchers are geeks .
Agrover112 Yeah, I'm afraid not many people will be interested. Even those videos which were requested don't get any views. Btw I don't do any special, I force myself to workout and I often procrastinate
@@OleksiiSidorov I understand
Why Keras is not so recommended to work in AI industry ?
It's quite outdated, and not so popular in general
for basic maths 3blue1brown channel
Unsure why it is important to read the classic papers.
Is this still relevant
Привет из России