Sneh Pandya

Postdoc @SkAI_Institute / @Fermilab | Physics PhD | Previously @Northeastern, @iaifi_news, @PhysicsIllinois & @NCSAatIllinois | ML x cosmology

[the finite physicist @FinitePhysicist]

Thanks to Terry Tao for featuring me on his blog!

https://t.co/CwtXD0jRfr

[Sneh Pandya @snehjp2]

I passed!

[Andrej Karpathy @karpathy]

@willcb Theoretical physicists are the intellectual embryonic stem cell, I’ve now seen them become ~everything.

[Sneh Pandya @snehjp2]

Last first day of school! (I think?)

[Sneh Pandya @snehjp2]

First time at @CERN! It has been a dream of mine for the past ~decade to visit here :)

[IAIFI @iaifi_news]

Our final lectures for the IAIFI Summer School were from @ACiprijanovic covering the “Domain Shift Problem: Building Robust AI Models with Domain Adaptation.”

[IAIFI @iaifi_news]

Shoutout to @snehjp2, this afternoon’s tutorial lead, who attended the first IAIFI Summer School and has been on the planning committee ever since, culminating in his role this year as a tutorial lead!

[Sneh Pandya @snehjp2]

Happy to share that our SIDDA paper is officially published in @MLSTjournal! I will be talking about this and domain adaptation more broadly at the @iaifi_news summer school tomorrow :)

https://t.co/D4tEwWbACJ

[Sneh Pandya @snehjp2]

New paper! Here, we introduce SIDDA: a semi-supervised, out-of-the-box domain adaptation method for classification, built upon optimal transport distances. It leverages existing computational resources (HT torch, geomloss) with minimal computational overhead.

[NSF-DOE Rubin Observatory @VRubinObs]

Introducing...your sneak peek at the cosmos captured by @NSF–@doescience Vera C. Rubin Observatory!

Can you guess what regions of sky they are?

This is just a peek...join us at 11am US EDT for your full First Look at how Rubin will #CaptureTheCosmos!

https://t.co/1a74X2edp8

[Claire Lamman @ClaireLamman]

DESI's DR2 BAO results are out!!
TL;DR... 1/n
@desisurvey

[Sneh Pandya @snehjp2]

Paper: https://t.co/4L2UNF9Ix1
Code: https://t.co/v1Z1BZSQjy
Data: https://t.co/PscK5X7rfF

with @purvikpatel_, @mike_walmsley_, @iamstarnord, @ACiprijanovic

Super thankful to work with this team of great people!

[Sneh Pandya @snehjp2]

New paper! Here, we introduce SIDDA: a semi-supervised, out-of-the-box domain adaptation method for classification, built upon optimal transport distances. It leverages existing computational resources (HT torch, geomloss) with minimal computational overhead.

[Sneh Pandya @snehjp2]

We benchmarked SIDDA on both simulated and real datasets, for a variety of shifts -- Poisson noise, PSF blurring, and galaxy images from different instruments. We also spent some time studying the inherent robustness of ENNs for classification tasks + motivating with DA theory.

[Sneh Pandya @snehjp2]

Moreover, SIDDA requires virtually no hyperparameter tuning (!!), a common struggle with existing DA methods, while inherently improving the calibration of NN-based classifiers. In many cases, it even increases performance on in-distribution data.

[Sneh Pandya @snehjp2]

SIDDA works by dynamically adjusting the transport plan and loss landscape during training for optimal domain alignment. It also performs particularly well when paired with ENNs, in some cases offering ~40% improvement in target data performance when compared to typical CNNs.

[Aleksandra Ciprijanovic @ACiprijanovic]

A cool new paper from my group. If your favorite ML model keeps performing poorly due to the domain shift, check our paper out. We demonstrate a fast, easy, and automated solution to domain adaptation with SIDDA!

https://t.co/ERIioWgH5V?

[IAIFI @iaifi_news]

Excited to announce that registration is open for our 2025 IAIFI Summer School! The Summer School will be held at Harvard, August 4–8, 2025 and offers lectures and hands-on tutorials for early career AI+Physics researchers. Apply by February 7, 2025: https://t.co/Re9YJarpEN

[the finite physicist @FinitePhysicist]

Physics is what physicists do