How to read research papers for machine learning?

( Advice from Andrew Ng, one of the most well known machine learning educators on the internet )

🧵👇
Step 1 : Collect relevant resources.

The internet is filled with information, blog posts, videos, GitHub repositories etc.

You are no machine, search for a few resources online, collect and organize them.

These could be medium posts, blog posts, GitHub repos etc.

(2 / 7)
Step 2 : Compile them to a list and skim through roughly 10% of each of the resources.

Let's say you like one of those papers a lot and you read through the entire thing, you might end up reading another research paper from the citations but maybe not the entire thing.

(3 / 7)
This way you will end up going through each resource in your list, maybe not a 100% but you'll have a rough idea about it. ( not to mention the extra papers you read through the citations and what not)

(4 / 7)
According to Andrew Ng, reading 5-20 papers will give you a basic understanding of subject, 50-100 will give you more than very good knowledge about it.

(5 / 7)
Step 3: How do you read a paper?

You could read it on your laptop, iPad or even physical, read in a way that works for you.

(Fun Fact : Andrew personally prefers physical papers)

(6 / 7)
I've spent over 3 hours on this thread and I've gotten a bunch of school work.😅

If you could support me by considering to follow me or liking my content, it would mean the world me, it takes a lot of effort to make this content. 🤩

Part 2 of this thread coming soon.😉

(7 / 7)
You can follow @PrasoonPratham.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled:

By continuing to use the site, you are consenting to the use of cookies as explained in our Cookie Policy to improve your experience.