Ramansh Sharma
@ramanshsharma.bsky.social
I do scientific machine learning research (operator learning at the moment). PhD student at the Kahlert School of Computing.
Reposted by Ramansh Sharma
Reposted by Ramansh Sharma
Reposted by Ramansh Sharma
hi everyone. i usually don’t do this but i’m struggling financially and it’s becoming a real burden especially with necessary expenses as a student (food, gas, rent, etc). if you enjoy my content at ALL and want to help out, my venmo is astrojaz.
you’re not obligated to give me anything of course 🥹
you’re not obligated to give me anything of course 🥹
February 24, 2025 at 1:40 AM
hi everyone. i usually don’t do this but i’m struggling financially and it’s becoming a real burden especially with necessary expenses as a student (food, gas, rent, etc). if you enjoy my content at ALL and want to help out, my venmo is astrojaz.
you’re not obligated to give me anything of course 🥹
you’re not obligated to give me anything of course 🥹
Reposted by Ramansh Sharma
Part 2 of the collaboration with Terence Tao on the cosmic distance ladder is now out. It covers how we first learned the distances to planets, stars, and galaxies far, far away.
youtu.be/hFMaT9oRbs4
youtu.be/hFMaT9oRbs4
The cosmic distance ladder with Terence Tao, part 2
YouTube video by 3Blue1Brown
youtu.be
February 23, 2025 at 3:00 PM
Part 2 of the collaboration with Terence Tao on the cosmic distance ladder is now out. It covers how we first learned the distances to planets, stars, and galaxies far, far away.
youtu.be/hFMaT9oRbs4
youtu.be/hFMaT9oRbs4
Reposted by Ramansh Sharma
Reposted by Ramansh Sharma
Reposted by Ramansh Sharma
I swear I'm not making a cooking channel here. But these palak puri came out so well.
January 19, 2025 at 8:01 PM
I swear I'm not making a cooking channel here. But these palak puri came out so well.
Reposted by Ramansh Sharma
SCI faculty members Amir Arzani, Kate Isaacs (@kisaacs.bsky.social), and Bei Wang Phillips recently received a prestigious national honor: the Presidential Early Career Award for Scientists and Engineers. Congrats to these rising stars! 🌟 #PECASE @utah.edu
Three SCI Faculty Earn Top National Award for Early Career Scientists and Engineers
www.sci.utah.edu
January 16, 2025 at 6:34 PM
SCI faculty members Amir Arzani, Kate Isaacs (@kisaacs.bsky.social), and Bei Wang Phillips recently received a prestigious national honor: the Presidential Early Career Award for Scientists and Engineers. Congrats to these rising stars! 🌟 #PECASE @utah.edu
Reposted by Ramansh Sharma
running vi inside a docker process running the dev build of the social app, ssh'd into a remote box from inside tmux running in alacritty, deleting javascript code as god intended
January 10, 2025 at 10:03 PM
running vi inside a docker process running the dev build of the social app, ssh'd into a remote box from inside tmux running in alacritty, deleting javascript code as god intended
Reposted by Ramansh Sharma
Reposted by Ramansh Sharma
20 years ago today, a tiny neutron star reached halfway across the galaxy and slammed our planet with a gamma-ray hammer.
www.scientificamerican.com/article/an-e...
🧪🔭
www.scientificamerican.com/article/an-e...
🧪🔭
An Earth-Rocking Cosmic Explosion Turns 20
Twenty years ago today a magnetar’s epic tantrum made our planet ring like a bell from tens of thousands of light-years away
www.scientificamerican.com
December 27, 2024 at 4:23 PM
20 years ago today, a tiny neutron star reached halfway across the galaxy and slammed our planet with a gamma-ray hammer.
www.scientificamerican.com/article/an-e...
🧪🔭
www.scientificamerican.com/article/an-e...
🧪🔭
I can do this.
December 21, 2024 at 5:28 AM
I can do this.
Reposted by Ramansh Sharma
ModernBERT is available as a slot-in replacement for any BERT-like model, with both 139M param and 395M param sizes.
It has a 8192 sequence length, is extremely efficient, is uniquely great at analyzing code, and much more. Read this for details:
huggingface.co/blog/modern...
It has a 8192 sequence length, is extremely efficient, is uniquely great at analyzing code, and much more. Read this for details:
huggingface.co/blog/modern...
December 19, 2024 at 4:45 PM
ModernBERT is available as a slot-in replacement for any BERT-like model, with both 139M param and 395M param sizes.
It has a 8192 sequence length, is extremely efficient, is uniquely great at analyzing code, and much more. Read this for details:
huggingface.co/blog/modern...
It has a 8192 sequence length, is extremely efficient, is uniquely great at analyzing code, and much more. Read this for details:
huggingface.co/blog/modern...
Reposted by Ramansh Sharma
Scientists discover NEW DEEPEST spot on Earth: Gukesh's opening prep. #DingGukesh
December 4, 2024 at 9:40 AM
Scientists discover NEW DEEPEST spot on Earth: Gukesh's opening prep. #DingGukesh
Reposted by Ramansh Sharma
Today along with my project co-lead @rubenohana.bsky.social and the team at @polymathicai.bsky.social I'm excited to announce the release of the Well, a 15TB collection of 15+ datasets for physical simulation.
Paper: openreview.net/pdf?id=00Sx5...
Github: github.com/PolymathicAI...
Paper: openreview.net/pdf?id=00Sx5...
Github: github.com/PolymathicAI...
December 2, 2024 at 5:33 PM
Today along with my project co-lead @rubenohana.bsky.social and the team at @polymathicai.bsky.social I'm excited to announce the release of the Well, a 15TB collection of 15+ datasets for physical simulation.
Paper: openreview.net/pdf?id=00Sx5...
Github: github.com/PolymathicAI...
Paper: openreview.net/pdf?id=00Sx5...
Github: github.com/PolymathicAI...
Reposted by Ramansh Sharma
Generating cat videos is nice, but what if you could tackle real scientific problems with the same methods? 🧪🌌
Introducing The Well: 16 datasets (15TB) for Machine Learning, from astrophysics to fluid dynamics and biology.
🐙: github.com/PolymathicAI...
📜: openreview.net/pdf?id=00Sx5...
Introducing The Well: 16 datasets (15TB) for Machine Learning, from astrophysics to fluid dynamics and biology.
🐙: github.com/PolymathicAI...
📜: openreview.net/pdf?id=00Sx5...
December 2, 2024 at 4:08 PM
Generating cat videos is nice, but what if you could tackle real scientific problems with the same methods? 🧪🌌
Introducing The Well: 16 datasets (15TB) for Machine Learning, from astrophysics to fluid dynamics and biology.
🐙: github.com/PolymathicAI...
📜: openreview.net/pdf?id=00Sx5...
Introducing The Well: 16 datasets (15TB) for Machine Learning, from astrophysics to fluid dynamics and biology.
🐙: github.com/PolymathicAI...
📜: openreview.net/pdf?id=00Sx5...
Residual connections are such a fascinating idea at their core! They approximate the same mapping that exists without the residual connection, but the residual connection helps training because of the inductive bias.
December 2, 2024 at 3:02 AM
Residual connections are such a fascinating idea at their core! They approximate the same mapping that exists without the residual connection, but the residual connection helps training because of the inductive bias.
Playing chess on a first date. Perchance
December 1, 2024 at 11:39 PM
Playing chess on a first date. Perchance
St George and Zion national park are beautiful!!
December 1, 2024 at 1:14 AM
St George and Zion national park are beautiful!!
Reposted by Ramansh Sharma
I’ve been hearing through the grapevine that the new SOAP optimizer might be SOTA now. Has anyone given it a go and compared it to Adam?
arxiv.org/abs/2409.11321
arxiv.org/abs/2409.11321
SOAP: Improving and Stabilizing Shampoo using Adam
There is growing evidence of the effectiveness of Shampoo, a higher-order preconditioning method, over Adam in deep learning optimization tasks. However, Shampoo's drawbacks include additional hyperpa...
arxiv.org
November 30, 2024 at 9:08 PM
I’ve been hearing through the grapevine that the new SOAP optimizer might be SOTA now. Has anyone given it a go and compared it to Adam?
arxiv.org/abs/2409.11321
arxiv.org/abs/2409.11321
Scientific and critical thinking, is both simple to do at any point, but difficult to master and enable as a constant back-of-the-mind process.
November 25, 2024 at 6:38 AM
Scientific and critical thinking, is both simple to do at any point, but difficult to master and enable as a constant back-of-the-mind process.
My mom made me a chef ❤️
November 23, 2024 at 4:20 AM
My mom made me a chef ❤️