Saurabh Patil

> Status: I am a Senior AI Research Engineer at Graylark Technologies, where I develop cutting-edge AI solutions for complex challenges.

Previously, I was a Research Assistant at Carnegie Mellon University (CMU). My research focused on Scientific Machine Learning (SciML), specifically developing Neural Operators and learning-based methods to accelerate traditional numerical simulations. I hold a Master's degree in Mechanical Engineering from CMU, where I bridged the gap between physics-based modeling and deep learning.

[Email] / [CV] / [Scholar] / [GitHub] / [LinkedIn] / [HuggingFace]
Saurabh Patil

Publications

* denotes equal contribution

CaFA

CaFA: Global Weather Forecasting with Factorized Attention on Sphere

Zijie Li, Anthony Zhou, Saurabh Patil, Amir Barati Farimani

arXiv, 2024

An end-to-end ML weather prediction model based on factorized attention on a sphere.

Paper
L-PDE

Latent Neural PDE Solver: a reduced-order modelling framework

Zijie Li*, Saurabh Patil*, Francis Ogoke, et al.

arXiv, 2024

Proposed learning system dynamics in latent space with coarser discretizations.

Paper
Hyena

Hyena Neural Operator for Partial Differential Equations

Saurabh Patil, Zijie Li, Amir Barati Farimani

APL Machine Learning, 2024

A state-space based neural operator with sub-quadratic complexity and global receptive field.

Code

Academic Projects

Point Clouds

Adversarial Attacks Against Point Clouds

Trustworthy AI Autonomy, 2022

Testing PointNet and DGCNN robustness against FGSM and Saliency map attacks.

Code
Tail Robot

Tail Robot-aeroDIMA

Bioinspired Robot Design, 2022

Comparing maneuverability of wheeled robots with aerodynamic vs. inertial tails.

Video | Poster