Vivan Rangra's Avatar

Vivan Rangra

@vivanrangra03.bsky.social

Aspiring statistician and ML researcher IIIT D

94 Followers  |  1,488 Following  |  12 Posts  |  Joined: 06.11.2024  |  1.5223

Latest posts by vivanrangra03.bsky.social on Bluesky

Dwarkesh Patel podcasts are super interesting especially if one's in AI .

01.12.2024 20:27 β€” πŸ‘ 4    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Rogan has diverse range of people even scientists like Brian Coxx and Neil Tyson for several times . I think its bad that we are putting a label onto the guy despite him being a LGBTQ ally and also a supporter of Bernie Sanders for serveral years .

01.12.2024 20:26 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Welcome @igilitschenski.bsky.social!

Your journey starts here πŸ‘‡
go.bsky.app/M7HGC3Y

25.11.2024 18:47 β€” πŸ‘ 8    πŸ” 3    πŸ’¬ 1    πŸ“Œ 0

Margin call

24.11.2024 12:04 β€” πŸ‘ 4    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

A good resource for Deep Learning in Graphs : yaoma24.github.io/dlg_book/dlg...
from Yao Ma and Jiliang Tang
#deeplearning #graphicalnn

23.11.2024 15:59 β€” πŸ‘ 1    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0

Only reason why I still have my PS2

23.11.2024 14:00 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

This is very important because there are many researchers here whose posts are informative and provide links that are worth reading later during free time.

23.11.2024 13:58 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

@support.bsky.team
Pls add a bookmark feature where we can add bookmarks people 's post and even have an option to categorize them[like in Instagram].

23.11.2024 13:55 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Preview
$100K or 100 Days: Trade-offs when Pre-Training with Academic Resources Pre-training is notoriously compute-intensive and academic researchers are notoriously under-resourced. It is, therefore, commonly assumed that academics can't pre-train models. In this paper, we seek...

A timely paper exploring ways academics can pretrain larger models than they think, e.g. by trading time against GPU count.

Since the title is misleading, let me also say: US academics do not need $100k for this. They used 2,000 GPU hours in this paper; NSF will give you that. #MLSky

23.11.2024 13:50 β€” πŸ‘ 144    πŸ” 12    πŸ’¬ 10    πŸ“Œ 3

Here is a list of ML OSS & Open Source / Science enthusiasts I found on Bluesky πŸ¦‹

go.bsky.app/8MFcfXd

Let me know if you find such people here!

I'm still new here and probably the list misses many must-add people, so let's built it togetherπŸ’ͺ

21.11.2024 05:19 β€” πŸ‘ 114    πŸ” 49    πŸ’¬ 41    πŸ“Œ 4
Post image

The birth of a star as seen by the James Webb Space Telescope

21.11.2024 08:03 β€” πŸ‘ 56143    πŸ” 3921    πŸ’¬ 445    πŸ“Œ 204

With increasingly tougher evaluations and the ambiguity of what the professor has taught and will ask in exams , more slide content is very helpful .
Some profs also don't attach readings from where they took topics/content, so a more in depth slides go long way.

21.11.2024 16:18 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

College ruined fun reading πŸ˜”

21.11.2024 16:15 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Congrats sir !

19.11.2024 21:29 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Converting them to pdf is a hassle but you can do it in bulk

19.11.2024 21:22 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

But it has a cutoff date and what if there have been changes in documentation?

19.11.2024 21:21 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

@vivanrangra03 is following 20 prominent accounts