iiinova — where your digital self keeps evolving

Arvind Mohan

@arvindmohan.bsky.social

169 followers 484 following 10 posts

Scientific ML for PDEs, Fluid Dynamics & Earth Sciences. Scientist @Los Alamos National Lab. Aerospace Engineer and Mountaineer. Opinions my own, not LANL/ US DOE

Posts Replies Media

Arvind Mohan @arvindmohan.bsky.social · 10/12/2024

At @agu.org with great colleagues and interesting work. And snarky badge stickers, courtesy of our very own US Dept of Energy 🙃

010

Arvind Mohan @arvindmohan.bsky.social · 02/12/2024

Or more accurately, discussing numerical methods for gradient descent while undergoing (rapid) gradient descent ⛷️

030

Reposted by Arvind Mohan

Jonah Miller @thephysicsmill.com · 29/11/2024

I'm really excited about this paper. Some context for 🔭🧪🔬 folks, as the AI summary may be a bit dry... A common activity in scientific ML is to train AI models on numerical data, generated by simulation. The simulation data is assumed to be ground truth... 🧵

1112

Reposted by Arvind Mohan

lebellig @lebellig.bsky.social · 28/11/2024

Cool article by @marccoru.bsky.social et al. exploring the use of spherical harmonics and very shallow SIREN networks to convert longitude and latitude meaningful geospatial embeddings on the sphere (code is also available) arxiv.org/abs/2310.06743

0132

Arvind Mohan @arvindmohan.bsky.social · 28/11/2024

This is a fantastic resource to make research more accessible!

0102

Arvind Mohan @arvindmohan.bsky.social · 26/11/2024

To add, even in cases where extrapolation is seen, its likely because of "lucky" interactions in numerical dissipation between the discretization scheme, the grid and the initial condition. Our paper provides formal tools to a priori estimate error and identify extrapolation limits *before* training

141

Arvind Mohan @arvindmohan.bsky.social · 26/11/2024

Just joined - Where the sky is still blue, but the bird is long gone.

030